8.7 Exercises for the CMS Data Analysis School (CMSDAS)

Complete: 5
Detailed Review status

ALERT! NOTE: - If you are not a facilitator or an organizer/responsible at a CMS Data Analysis School, kindly do not edit this page.

Introduction

CMS is a big collaboration of three thousand people spread over several institutions across five continents. It is very important to bring the collaborators up to speed in terms of learning about the CMS detector and computing software in order to contribute towards the physics analysis. CMS maintains the WorkBook and SWGuide as much as up-to-date as possible to help guide about every aspect of the CMS detector. In addition to it, several tutorials and workshops are held regularly to give a hands on experience to the users. There are tutorials on PAT (Physics Analysis Toolkit), CRAB, RooStats, Fireworks, Python in CMSSW etc. where one learns the basic CMS software tools. It is also pertinent that people are able to apply these tools. With this goal in mind, an Extended-JTerm (EJTerm) workshop was held at the LPC (LHC Physics Center) at Fermilab from 5-10 Jan, 2010, to help CMS physicists learn about CMS commissioning and analysis and thereby participate in significant ways to the discovery and elucidation of the new physics. Several institution contributed in organizing and preparing for this workshop. The EJTerm workshop model was very successful. To encompass yet a bigger role, the name EJTerm was changed to CMS Data Analysis School (CMSDAS) in 2011. The very first school with the new name CMSDAS was held at LPC, Fermilab in 25-29 Jan 2011. Since then CMSDAS has been held a several places worldwide.

Links to all the past and upcoming CMS Schools can be found HERE.

This twiki explains the concept of these schools and summarizes the exercises held during it. It also gives links to the actual exercises. The exercises get updated regularly before an upcoming CMSDAS.

Contents

The CMS Data Analysis School

The concept of CMS Schools was born in 2010 with the idea to build a sustainable collaboration where young member physicists can reach their full potential by training, engage and nurturing them in all aspects of experimental physics - data analysis, software and hardware, and career development paths. The CMS Schools are based on experiential learning and 90% hands-on and 10% lectures. It started as physics analysis school called CMSDAS (CMS Data Analysis School). We have expanded this idea train students in upgrade hardware. This efforts is called CMS Upgarde School or CUPS for short. The school introduces new members of our collaboration, students, post docs and faculty, to CMS software and available tools. It provides a unique opportunity to get a hands-on experience with physics analysis and detector hardware and upgrades. To prepare the participants, a series of pre-workshop online activities “extended-tutorials” are prepared in the form of exercises. Examples include the Physics Analysis Toolkit, the Fireworks event display, and the CRAB/DAS system to locate and transfer data. CRAB means CMS Remote Analysis Builder and DAS here means Data Agrgegation Service. Several CMS institutions worked with those who signed up for CMSDAS to ensure that, when they arrive for school in January, they have already attained a working knowledge of the indispensable tools needed to confront the first CMS data.

During the week of the school, participants work in small teams on examples of collision data analysis. The problems ranged from the simple: measuring the charged track multiplicity in early data, to commissioning exercises, and simulated physics measurements such as measuring the Y(nS) cross section and searches for new physics. The teams are facilitated by the experienced volunteers from the CMS community in US and Europe. Teams present their physics measurements in a “conference” at the end of the school.

Links to all the past and upcoming CMS Schools can be found HERE.

Guidelines to prepare SHORT and LONG exercises

To maintain the exercise and software for it, in the long term, you should write your exercise in whatever format you want to use (EDAnalyzers, bare TTrees, text dump, whatever), HOWEVER there are only a limited number of tools that you can use IF you want your software to make it into CMSSW as an official example for long-term maintenance by the offline group.

The list of things that you can use for this purpose are:

  • EDAnalyzers/EDFilters/EDProducers
  • FWLite software in any form
  • Python configuration to CMSSW or FWLite
  • PYROOT in FWLite
As long as your software uses those above, you can use any object that you desire (reco:: versions, pat:: versions, bare objects like doubles and ints, etc, etc, etc) and the software you write will automatically be maintainable.

Please be aware that you're free to opt out of this service!

Furthermore, there is a 100% one-to-one corresponding structure in CMSSW that is exactly the same functionality as a bare TTree, and this is writing bare native types (doubles, ints, etc) directly into an EDM-file (which, of course, is a TTree at the end). This is only a syntactical change, and not a semantic change. The small initial burden of doing this is actually quite small from experience, and it is strongly recommend making the switch to the EDM-compliant TTree structures.

  • Naming convention for the SHORT and LONG exercises twiki - Please have the name of your exercise pre-fixed with https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2015XXXXX. For example https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2015HiggsSearchExercise etc.

  • If you are working on your SHORT or LONG exercise, please make a RED flag against it showing its status on this page ( worked on, not ready etc.). Also please feel free to write yourself about your exercise just like in the previous CMSDAS 2011 exercises below.

CMSDAS 2016 (LPC, NTU/Taipei, Hamburg)

Pre-Workshop Exercises

Short Exercises in Hamburg

1. Generators: SWGuideCMSDataAnalysisSchoolHamburg2016GeneratorExercise

Description: An introduction to the physics of datasets and an exploration of Monte Carlo truth information.


Responsible at CMSDAS@Hamburg : Alexis Kalogeropoulos (DESY), Thomas Peiffer (UniHH)

2. Pileup and MET : SWGuideCMSDataAnalysisSchoolHamburg2016PileupAndMETExercise

Description: The goal of this exercise is to familiarize you with the tools available for measuring and mitigating pileup in an event. To this end, we will be discussing; where pileup comes from, how we simulate pileup for MC, and how it is measured in data. There are several variables that are strongly correlated to pileup and we will discuss the merits of each as well as how to access them. Additionally, we will be covering MET and the corrections made to MET in response to pileup corrections made to jets. Finally, this course will cover some new tools for calculating muon isolation and how this relates to pileup.

For each of these topics we will be discussing run 2 methods, which can be implemented on the MiniAOD data tier.


Responsible at CMSDAS@Hamburg: Benedikt Vormwald (UHH), Daniel Troendle (UHH)

3. Electrons+Photons: SWGuideCMSDataAnalysisSchoolHamburg2016ElectronsAndPhotons

Description: This short exercise goes through the steps of accessing and displaying kinematic variables for electrons and photons, using ID criteria to improve signal efficiency and background rejection, and creating and improving a dielectron invariant mass distribution.


Responsible at CMSDAS@Hamburg : Knut Kiesel, Christian Schomakers (ACIB), Swagata Mukherjee (ACIIIA)

4. Muons: SWGuideCMSDataAnalysisSchoolHamburg2016Muons

Description: This exercise has the goal to familiarize you with muon identification and isolation in CMS. It starts with a presentation describing the muon reconstruction and ID in CMS and then some interactive exercises are being performed to demonstrate the details. The first interactive part of this exercise is to learn to use muons, muon discriminators and muon isolation and to be able to choose the best recipe depending on the analysis you want to use. The exercise will conclude with a discussion on the measurement of muon efficiencies and the application of muon calibration and scale corrections.


Responsible at CMSDAS@Hamburg : Francesco Costanza (DESY), Elizabeth Kennedy (UCR)

5. Tracking and Primary Vertices: SWGuideCMSDataAnalysisSchoolHamburg2016TrackingAndVertexingExercise

Description: We will present an introduction to using tracks for analyses in the era of large pile-up (many primary vertices). Our exercises will all use real data and will familiarize you with the following techniques:

  • Extracting basic track parameters and reconstructing invariant masses from tracks in CMSSW. Tracks are the closest detector entity to the four-vectors of particles: the momentum of a track is nearly the momentum of the charged particle itself.
  • Extracting basic parameters of primary vertices. In this high-luminosity era, it is not uncommon for a single event to contain as many as ten to twenty independent collisions. For most analyses, only one is relevant, and it can usually be identified by its tracks.
  • Reconstructing secondary vertices to measure the Kshort mass


Responsible at CMSDAS@Hamburg : Gregor Mittag (DESY)

6. Visualization and Event Display: SWGuideCMSDataAnalysisSchoolHamburg2016EventDisplayExercise

Description: In this short exercise you learn how to explore interesting events using a physics event display. Event displays help understand the physics processes and particle interactions you are investigating. They are also a great tool to scan the detector for performance issues. Physics objects, like Electrons, Muons, Jets and more, can be visualized along with the various sub-detector systems. We will use Fireworks, the CMS event-display, to scan through a number of 13TeV Monte-Carlo- and data-events and in order to produce some nice snapshots of the CMS detector in action.


Responsible at CMSDAS@Hamburg : Swagata Mukherjee (ACIIIA), Heiner Tholen (UHH)

7. Trigger: SWGuideCMSDataAnalysisSchoolHamburg2016TriggerExercise

Description: This exercise is intended to familiarize you with finding trigger related information relevant for your physics analysis, accessing them from within your analyzer in view of using them to estimate the trigger efficiency for any given analysis. In the final discussion we will use what learned in the exercise and try to elaborate about data driven methods to estimate the trigger efficiency, or to provide correction factors to the MC one, and systematic effects associated to the possible methods chosen to compute such trigger efficiency.


Responsible at CMSDAS@Hamburg : TBD

8. B Tagging: SWGuideCMSDataAnalysisSchoolHamburg2016BTaggingExercise

Description: This exercise provides a pedagogical introduction to b tagging and a hands-on experience accessing and plotting b-tag discriminators for various b tagging algorithms. Furthermore, students will learn how to plot the b-tagging efficiency and the mistag rate as a function of jet kinematics as well as how to compare the performance of different b tagging algorithms. Finally, the exercise will provide a comprehensive reference to more advanced topics and additional resources.


Responsible at CMSDAS@Hamburg : Marek Nedziela (UHH)

9. Statistics: SWGuideCMSDataAnalysisSchoolHamburg2016StatisticsExercise

Description: In these exercises you will learn how to use the statistical modeling toolkit RooFit and the associated statistical analysis toolkit RooStats to perform some standard, but very widely used, statistical procedures, including fitting, estimating (i.e., measuring) parameters, computing intervals, and setting limits. You will also learn how to use the tool combine, which, as its name suggests, can be used to perform statistical analyses of multiple results.


Responsible at CMSDAS@Hamburg : Adrian Perieanu (UHH), Jory Sonneveld (UHH)

10. Taus: SWGuideCMSDataAnalysisSchoolHamburg2016TauExercise

Description: Jet to tau fake rate measurement in data and MC simulation using 13 TeV data samples. These measurements will include both measurement of the gluon jet fake rate (usually from multi-jet QCD events) and quark jet fake rate (from the jets which are accompanied by W boson, i.e. in W+Jet events). Lesson to be learned: Participants will learn how a tau candidate is reconstructed in CMS and what are the main properties of the hadronic tau decays. They will learn about possibility of the jet to be mis-identified as a tau candidate and measure this probability (jet-to-tau fake rate) and can measure this rate in terms of the several quantities like tau-Pt, tau-Eta,tau-Phi and number of reconstructed vertices in the events. Additionally they will be able to compare the tau fake for gluon and quark jet and compare the fake rate in data and MC.


Responsible at CMSDAS@Hamburg : Adrian Perieanu(UHH), Alexei Raspereza (DESY)

11. Jets: SWGuideCMSDataAnalysisSchoolHamburg2016JetExercise

Description:
  • A 101 on how to access jets in the CMS framework without assuming prior knowledge of jet analysis.
  • Make you familiar with basic jet types and algorithms and how to use them in your analysis.
  • Illustrate each exercise using real life example scripts.
  • Give you a comprehensive reference to more advanced workbook examples, additional resources, and pedagogical documentation in one place.


Responsible at CMSDAS@Hamburg : Natalia Kovalcuk (UHH), Fred Stober (UHH)

Long Exercises in Hamburg

1. Top quark mass measurement : SWGuideCMSDataAnalysisSchoolHamburg2016TopMassExercise

Description: Students will learn how to measure the top quark mass from the latest 13 TeV data using b-jets and simple two-body kinematics. The analysis will include proper treatment of backgrounds and systematic uncertainties involved in this analysis with collision data. The analysis performed at this school is likely to be the first top quark measurement ever at 13 TeV, and students are invited to continue this analysis all the way to a public result in the Spring of 2016, as part of the larger CMS analysis team. The kiematical details of the this method and idea are given at the following link http://arxiv.org/abs/1209.0772

Recommended Pre-requisite Short Exercises: Statistics, Jets, b-tagging


Responsible at CMSDAS@Hamburg : Altan Cakir (ITU), Alexander Grohsjean (DESY), Fred Stober (UHH)

2. ZprimeToDiLeptons : SWGuideCMSDataAnalysisSchoolHamburg2016ZprimeDiLeptonsExercise

Description: The Zprime to dilepton exercise focusses on the search for heavy resonances decaying to muon pairs using data collected in 2015 at 13 TeV, with the CMS detector. The students wll become familiar with the analysis strategy optimized for the Run2 LHC data. Data-driven methods used to determine the detector resolution and background composition, as well as predictions derived from simulation, will be described and tested. Results will discussed in term of the final invariant mass plots and with the exclusion limits on the Zprime mass for several new Physics Models.

Recommended Pre-requisite Short Exercises: Muon, Generator, Statistics


Responsible at CMSDAS@Hamburg : Elizabeth Kennedy (UCR)

3. Heavy Higgs Searches : SWGuideCMSDataAnalysisSchoolHamburg2016HeavyHiggsSearchesExercise

Description: This analysis describes search for new physics in photon + MET final state. This exercise is based mainly on data collected for Run2. Students will perform the efficiency measurements for the selection, background estimation, control region studies to cross check the major background estimates (Zgamma and Wgamma) and then set the limits on dark matter and extra dimension models.

Recommended Pre-requisite Short Exercises: Electrons/Photons, Statistics, MET


Responsible at CMSDAS@Hamburg : Benedikt Vormwald (UHH), Daniel Troendle (UHH)

4. Inclusive jets and inclusive b jets : SWGuideCMSDataAnalysisSchoolHamburg2016InclusiveJetsExercise

Description: This analysis is a generic search for new physics. It is motivated by .... This exercise will follow the recent analysis of the 2.3 fb-1 of data collected in 2015 at a center-of-mass energy of 13 TeV. The focus of the exercise will be on the ....

Recommended Pre-requisite Short Exercises: Jets, PileUp /MET and statistics short exercises


Responsible at CMSDAS@Hamburg : Paolo Gunnellini (DESY), Benoit Roland (DESY)

5. B2G Boosted Z'->ttbar semileptonic : SWGuideCMSDataAnalysisSchoolHamburg2016B2GTTbarExercise

Description: The objective of this exercise is to perform a search for top-antitop resonances in the 13 TeV LHC Run 2. Students will perform an efficiency measurement for signal MC, optimize event selections to maximize S/sqrt(B), investigate top and W tagging algorithms, estimate the backgrounds, and perform a likelihood minimization to search for new physics in the top-antitop spectrum.

Recommended Pre-requisite Short Exercises: Statistics, Tracking/Vertexing, Electron/Photon, Muon, Jets, b-tagging short exercises


Responsible at CMSDAS@Hamburg : Thomas Peiffer (UHH), Nataliia Kovalcuk (UHH), Marek Nedziela (UHH), Dominik Nowatschin (UHH)

6. Higgs->bb : SWGuideCMSDataAnalysisSchoolHamburg2016HiggsbbbarExercise

Description: The observation of a Higgs-like boson with a mass of 125 GeV was an extraordinary event for the CMS experiment and for particle physics as a whole. We must now characterize the properties of this new boson and confirm or reject the hypothesis that it is the Standard Model (SM) Higgs boson. One fundamental property to study is the new boson's coupling to fermions. Do fermions receive mass through the SM Higgs mechanism? We will focus on the search for the H->bb decay, which has the largest branching ratio (~60% at 125 GeV). The search is conducted in the final state where the Higgs boson (H) is produced in association with a vector boson (V = W or Z). This is commonly referred to as the VH(bb) analysis. In this exercise, we will start with an overview of the differences between signal and background processes and the corresponding discriminating variables. We will define our selection criteria in order to maximize our sensitivity to signal and use control regions in the data to estimate the backgrounds. We will perform a fit using the dijet invariant mass as the discriminant to observe an excess or set upper limits on the Higgs boson production cross section.

Recommended Pre-requisite Short Exercises:


Responsible at CMSDAS@Hamburg : Gregor Mittag (DESY) , Rostyslav Shevchenko (DESY), Roberval Walsh (DESY)

7. Z to tau tau cross-section measurement at 13TeV : SWGuideCMSDataAnalysisSchoolHamburg2016LongExerciseTauExercise

Description: The aim is to measure Z to tau tau cross-section @ 13 TeV. What we plan to do is select Z -> tau tau -> mu tau events both in data (single muon dataset) and MC (DYJets as signal and QCD, WJet and TTbar, and possibily diboson as backgrounds) Subtract all background from data. The difference between data and background will be considered as Z to tau tau. All backgrounds are estimated from MC, except QCD, which is computed from data using the mu-tau same-sign control region.

Recommended Pre-requisite Short Exercises: Tau


Responsible at CMSDAS@Hamburg : Francesco Costanza (DESY), Adrian Perieanu(UHH), Alexei Raspereza (DESY)

8. SUSY Edge Search : SWGuideCMSDataAnalysisSchoolHamburg2016SUSYEdgeSearchExercise

Description: The goal of this exercise is to perform a search for new physics with same-sign dilepton final state. This final state could manifest itself in a range of scenarios: supersymmetry particles decay, Majorana neutrinos, top quark partners with charge 5/3 and so on. The example signal to be used in this exercise is gluino-pair production with each gluino decaying to a top-quark pair and a stable neutralino. But the overall analysis design is kept general in order to be open to other possibilities. The main focus of the exercise is on how to estimate the standard model backgrounds in this search, and how to design a search strategy in order to maximize sensitivity to possible new physics scenarios.

Recommended Pre-requisite Short Exercises: Electrons/Photons, Muons, Statistics


Responsible at CMSDAS@Hamburg : Knut Kiesel (ACIB), Christian Schmökers (ACIB), Claudia Seitz (DESY)

9. Exotica Displaced Vertices : SWGuideCMSDataAnalysisSchoolHamburg2016ExoticaDisplacedVertices

Description: In this exercise, you will perform a search for a heavy resonance (M) decaying to two long-lived massive neutral particles (m), which then decay to leptons (electrons and muons). The full chain is M -> m m -> 4 leptons, with both pairs of leptons being displaced from the primary vertex. This highly exotic signature would be clear evidence of physics beyond the Standard Model, and is motivated by hidden valley scenarios.

You will perform the complete analysis with full statistics. The exercise will cover MC simulation, kinematics, definition of acceptance, efficiencies, and will use the RooStats tools to set upper limits (unless we see something new). We will discuss and explore sources of systematic uncertainty and how to estimate them using data-driven methods wherever possible. Participants will have the opportunity to be the first to explore this parameter space in the Run 2 data, since there is no active analysis targeting this final state.

Recommended Pre-requisite Short Exercises: Tracking + Primary Vertices, Muons, Electrons + Photons


Responsible at CMSDAS@Hamburg : Swagata Mukherjee (AC IIIA), Jory Sonneveld (UHH)

Short Exercises at NTU, Taipei

This is a template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT TO DESCRIBE YOUR SHORT EXERCISE/FACILITATORS IN THIS SECTION BELOW) - NameOfShortExercise : twiki url

Description: CUT PASTE DESCRIPTION HERE


Responsible at CMSDAS@NTU, Taipei : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

1. Generators: SWGuideCMSDataAnalysisSchool2016GeneratorExerciseatNTU

Description: An introduction to the physics of datasets and an exploration of Monte Carlo truth information.


Responsible at CMSDAS@NTU, Taipei : Vieri Candelise / Shin-Shan Yu

2. Pileup and MET : SWGuideCMSDataAnalysisSchoolNTU2016PileupAndMETExercise

Description: The goal of this exercise is to familiarize you with the tools available for measuring and mitigating pileup in an event. To this end, we will be discussing; where pileup comes from, how we simulate pileup for MC, and how it is measured in data. There are several variables that are strongly correlated to pileup and we will discuss the merits of each as well as how to access them. Additionally, we will be covering MET and the corrections made to MET in response to pileup corrections made to jets. Finally, this course will cover some new tools for calculating muon isolation and how this relates to pileup.

For each of these topics we will be discussing run 2 methods, which can be implemented on the MiniAOD data tier.


Responsible at CMSDAS@NTU, Taipei : Yueh-Feng Liu

3. Electrons+Photons: SWGuideCMSDataAnalysisSchoolNTU2016ElectronsAndPhotons

Description: This short exercise goes through the steps of accessing and displaying kinematic variables for electrons and photons, using ID criteria to improve signal efficiency and background rejection, and creating and improving a dielectron invariant mass distribution.


Responsible at CMSDAS@NTU, Taipei : Yu Chul Yang / Chia-Ming Kuo

4. Muons: SWGuideCMSDataAnalysisSchoolNTU2016Muons

Description: This exercise has the goal to familiarize you with muon identification and isolation in CMS. It starts with a presentation describing the muon reconstruction and ID in CMS and then some interactive exercises are being performed to demonstrate the details. The first interactive part of this exercise is to learn to use muons, muon discriminators and muon isolation and to be able to choose the best recipe depending on the analysis you want to use. The exercise will conclude with a discussion on the measurement of muon efficiencies and the application of muon calibration and scale corrections.


Responsible at CMSDAS@NTU, Taipei : Junghwan John Goh / Geonmo Ryu

5. Tracking and Primary Vertices: SWGuideCMSDataAnalysisSchoolNTU2016TrackingAndVertexingExercise

Description: We will present an introduction to using tracks for analyses in the era of large pile-up (many primary vertices). Our exercises will all use real data and will familiarize you with the following techniques:

  • Extracting basic track parameters and reconstructing invariant masses from tracks in CMSSW. Tracks are the closest detector entity to the four-vectors of particles: the momentum of a track is nearly the momentum of the charged particle itself.
  • Extracting basic parameters of primary vertices. In this high-luminosity era, it is not uncommon for a single event to contain as many as ten to twenty independent collisions. For most analyses, only one is relevant, and it can usually be identified by its tracks.
  • Reconstructing secondary vertices to measure the Kshort mass


Responsible at CMSDAS@NTU, Taipei : Tamas Almos VAMI

6. Visualization and Event Display: SWGuideCMSDataAnalysisSchoolNTU2016EventDisplayExercise

Description: In this short exercise you learn how to explore interesting events using a physics event display. Event displays help understand the physics processes and particle interactions you are investigating. They are also a great tool to scan the detector for performance issues. Physics objects, like Electrons, Muons, Jets and more, can be visualized along with the various sub-detector systems. We will use Fireworks, the CMS event-display, to scan through a number of 13TeV Monte-Carlo- and data-events and in order to produce some nice snapshots of the CMS detector in action.


Responsible at CMSDAS@NTU, Taipei : Yuan CHAO

7. Trigger: SWGuideCMSDataAnalysisSchoolNTU2016TriggerExercise

Description: This exercise is intended to familiarize you with finding trigger related information relevant for your physics analysis, accessing them from within your analyzer in view of using them to estimate the trigger efficiency for any given analysis. In the final discussion we will use what learned in the exercise and try to elaborate about data driven methods to estimate the trigger efficiency, or to provide correction factors to the MC one, and systematic effects associated to the possible methods chosen to compute such trigger efficiency.


Responsible at CMSDAS@NTU, Taipei : Phat S.

8. B Tagging: SWGuideCMSDataAnalysisSchoolNTU2016BTaggingExercise

Description: This exercise provides a pedagogical introduction to b tagging and a hands-on experience accessing and plotting b-tag discriminators for various b tagging algorithms. Furthermore, students will learn how to plot the b-tagging efficiency and the mistag rate as a function of jet kinematics as well as how to compare the performance of different b tagging algorithms. Finally, the exercise will provide a comprehensive reference to more advanced topics and additional resources.


Responsible at CMSDAS@NTU, Taipei : Andrey Pozdnyakov

9. Statistics: SWGuideCMSDataAnalysisSchoolNTU2016Statistics

Description: In these exercises you will learn how to use the statistical modeling toolkit RooFit and the associated statistical analysis toolkit RooStats to perform some standard, but very widely used, statistical procedures, including fitting, estimating (i.e., measuring) parameters, computing intervals, and setting limits. You will also learn how to use the tool combine, which, as its name suggests, can be used to perform statistical analyses of multiple results.


Responsible at CMSDAS@NTU, Taipei : Mingshui Chen / Shilpi Jain

10. PPD and data preparation: SWGuideCMSDataAnalysisSchoolNTU2016PPDExercise

Description: In this set of labs you will learn how to look for datasets and find out their key properties relevant for analysis: how to navigate their parent-child relationship and determine the software release, production configuration and alignment-calibration conditions they've been produced with. You'll be exposed to using the principal web services providing status and details of datasets: das, McM, pMp and the DQM g.u.i. You'll also find out how to compute the integrated luminosity for your analysis.


Responsible at CMSDAS@NTU, Taipei : Giovanni Franzoni

11. Taus: SWGuideCMSDataAnalysisSchoolNTU2016TauExercise

Description: Jet to tau fake rate measurement in data and MC simulation using 13 TeV data samples. These measurements will include both measurement of the gluon jet fake rate (usually from multi-jet QCD events) and quark jet fake rate (from the jets which are accompanied by W boson, i.e. in W+Jet events). Lesson to be learned: Participants will learn how a tau candidate is reconstructed in CMS and what are the main properties of the hadronic tau decays. They will learn about possibility of the jet to be mis-identified as a tau candidate and measure this probability (jet-to-tau fake rate) and can measure this rate in terms of the several quantities like tau-Pt, tau-Eta,tau-Phi and number of reconstructed vertices in the events. Additionally they will be able to compare the tau fake for gluon and quark jet and compare the fake rate in data and MC.


Responsible at CMSDAS@NTU, Taipei : TBD

12. Jets: SWGuideCMSDataAnalysisSchoolJetAnalysisNTU2016

Description:
  • A 101 on how to access jets in the CMS framework without assuming prior knowledge of jet analysis.
  • Make you familiar with basic jet types and algorithms and how to use them in your analysis.
  • Illustrate each exercise using real life example scripts.
  • Give you a comprehensive reference to more advanced workbook examples, additional resources, and pedagogical documentation in one place.


Responsible at CMSDAS@NTU, Taipei : TBD

Long Exercises at NTU, Taipei

This is a template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT TO DESCRIBE YOUR LONG EXERCISE/FACILITATORS IN THIS SECTION BELOW) - NameOfLongExercises : twiki url

Description: CUT PASTE DESCRIPTION HERE

Recommended Pre-requisite Short Exercises: For example Electrons, b-tagging etc.


Responsible at CMSDAS@NTU, Taipei : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

1. Top quark mass measurement from b jets and simple two-body kinematics : SWGuideCMSDataAnalysisSchoolNTU2016TopExercise

Description: Students will learn how to measure the top quark mass from the latest 13 TeV data using b-jets and simple two-body kinematics. The analysis will include proper treatment of backgrounds and systematic uncertainties involved in this analysis with collision data. The analysis performed at this school is likely to be the first top quark measurement ever at 13 TeV, and students are invited to continue this analysis all the way to a public result in the Spring of 2016, as part of the larger CMS analysis team. The kiematical details of the this method and idea are given at the following link http://arxiv.org/abs/1209.0772

Recommended Pre-requisite Short Exercises: Statistics, Jets, b-tagging


Responsible at CMSDAS@NTU, Taipei : Junghwan John Goh / Geonmo Ryu

2. ZprimeToDiLeptons : SWGuideCMSDataAnalysisSchoolNTU2016ZprimeDiLeptons

Description: The Zprime to dilepton exercise focusses on the search for heavy resonances decaying to muon pairs using data collected in 2015 at 13 TeV, with the CMS detector. The students wll become familiar with the analysis strategy optimized for the Run2 LHC data. Data-driven methods used to determine the detector resolution and background composition, as well as predictions derived from simulation, will be described and tested. Results will discussed in term of the final invariant mass plots and with the exclusion limits on the Zprime mass for several new Physics Models.

Recommended Pre-requisite Short Exercises: Muon, Generator, Statistics


Responsible at CMSDAS@NTU, Taipei : Vieri Candelise / Yu Chul Yang

3. Dark Matter mono-photons : https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolNTU2016MonophotonDM

Description: This analysis describes search for new physics in photon + MET final state. This exercise is based mainly on data collected for Run2. Students will perform the efficiency measurements for the selection, background estimation, control region studies to cross check the major background estimates (Zgamma and Wgamma) and then set the limits on dark matter and extra dimension models.

Recommended Pre-requisite Short Exercises: Electrons/Photons, Statistics, MET


Responsible at CMSDAS@NTU, Taipei : Chia-Ming Kuo / Shilpi Jain

4. SUSY hadronic: multijets+MET : https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolNTU2016SUSYHadronic

Description: This analysis is a generic search for new physics. It is motivated by models of R-parity conserving Supersymmetry (SUSY) that generally assume the existence of additional elementary particles. The SUSY signature we are searching for in the detector consists of several jets with high pT, large missing transverse momentum due to the LSPs, and no leptons. The data-based determination of the SM backgrounds is one of the key features of this analysis. This exercise will follow the recent analysis of the 2.1 fb-1 of data collected in 2015 at a center-of-mass energy of 13 TeV. The focus of the exercise will be on the background determination methods which are the heart of this analysis as well as the interpretation of the search results for SUSY models.

Recommended Pre-requisite Short Exercises: Jets, PileUp /MET and statistics short exercises


Responsible at CMSDAS@NTU, Taipei : Yueh-Feng Liu / Phat S.

5. B2G Boosted Z'->ttbar semileptonic : SWGuideCMSDataAnalysisSchoolNTU2016B2GTTbar

Description: The objective of this exercise is to perform a search for top-antitop resonances in the 13 TeV LHC Run 2. Students will perform an efficiency measurement for signal MC, optimize event selections to maximize S/sqrt(B), investigate top and W tagging algorithms, estimate the backgrounds, and perform a likelihood minimization to search for new physics in the top-antitop spectrum.

Recommended Pre-requisite Short Exercises: Statistics, Tracking/Vertexing, Electron/Photon, Muon, Jets, b-tagging short exercises


Responsible at CMSDAS@NTU, Taipei : Shin-Shan Yu / Yuan CHAO

6. Higgs->bb : https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolNTU2016Higgsbbbar2016Exercise

Description: The observation of a Higgs-like boson with a mass of 125 GeV was an extraordinary event for the CMS experiment and for particle physics as a whole. We must now characterize the properties of this new boson and confirm or reject the hypothesis that it is the Standard Model (SM) Higgs boson. One fundamental property to study is the new boson's coupling to fermions. Do fermions receive mass through the SM Higgs mechanism? We will focus on the search for the H->bb decay, which has the largest branching ratio (~60% at 125 GeV). The search is conducted in the final state where the Higgs boson (H) is produced in association with a vector boson (V = W or Z). This is commonly referred to as the VH(bb) analysis. In this exercise, we will start with an overview of the differences between signal and background processes and the corresponding discriminating variables. We will define our selection criteria in order to maximize our sensitivity to signal and use control regions in the data to estimate the backgrounds. We will perform a fit using the dijet invariant mass as the discriminant to observe an excess or set upper limits on the Higgs boson production cross section.

Recommended Pre-requisite Short Exercises: Statistics, b-tagging, Muon, Electron


Responsible at CMSDAS@NTU, Taipei : Andrey Pozdnyakov / Mingshui Chen

7. Exotica Displaced Vertices : SWGuideCMSDataAnalysisSchoolNTU2016ExoticaDisplacedVertices

Description: In this exercise, you will perform a search for a heavy resonance (M) decaying to two long-lived massive neutral particles (m), which then decay to leptons (electrons and muons). The full chain is M -> m m -> 4 leptons, with both pairs of leptons being displaced from the primary vertex. This highly exotic signature would be clear evidence of physics beyond the Standard Model, and is motivated by hidden valley scenarios.

You will perform the complete analysis with full statistics. The exercise will cover MC simulation, kinematics, definition of acceptance, efficiencies, and will use the RooStats tools to set upper limits (unless we see something new). We will discuss and explore sources of systematic uncertainty and how to estimate them using data-driven methods wherever possible. Participants will have the opportunity to be the first to explore this parameter space in the Run 2 data, since there is no active analysis targeting this final state.

Recommended Pre-requisite Short Exercises: Tracking + Primary Vertices, Muons, Electrons + Photons


Responsible at CMSDAS@NTU, Taipei : TBD

8. Z to tau tau cross-section measurement at 13TeV : SWGuideCMSDataAnalysisSchoolNTU2016LongExerciseTau

Description: The aim is to measure Z to tau tau cross-section @ 13 TeV. What we plan to do is select Z -> tau tau -> mu tau events both in data (single muon dataset) and MC (DYJets as signal and QCD, WJet and TTbar, and possibily diboson as backgrounds) Subtract all background from data. The difference between data and background will be considered as Z to tau tau. All backgrounds are estimated from MC, except QCD, which is computed from data using the mu-tau same-sign control region.

Recommended Pre-requisite Short Exercises: Tau


Responsible at CMSDAS@NTU, Taipei : TBD

9. SUSY Same-sign dilepton searches : SWGuideCMSDataAnalysisSchoolNTU2016SameSignDileptons

Description: The goal of this exercise is to perform a search for new physics with same-sign dilepton final state. This final state could manifest itself in a range of scenarios: supersymmetry particles decay, Majorana neutrinos, top quark partners with charge 5/3 and so on. The example signal to be used in this exercise is gluino-pair production with each gluino decaying to a top-quark pair and a stable neutralino. But the overall analysis design is kept general in order to be open to other possibilities. The main focus of the exercise is on how to estimate the standard model backgrounds in this search, and how to design a search strategy in order to maximize sensitivity to possible new physics scenarios.

Recommended Pre-requisite Short Exercises: Electrons/Photons, Muons, Statistics


Responsible at CMSDAS@NTU, Taipei : TBD

10. B meson cross section : SWGuideCMSDataAnalysisSchoolNTU2016BMesonCrossSection

Description: TBA

Recommended Pre-requisite Short Exercises: Tracking/PV, Muons, Statistics


Responsible at CMSDAS@NTU, Taipei : Sanjay Swain

Short Exercises at LPC

This is a template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT TO DESCRIBE YOUR SHORT EXERCISE/FACILITATORS IN THIS SECTION BELOW) - NameOfShortExercise : twiki url

Description: CUT PASTE DESCRIPTION HERE


Responsible at CMSDAS@LPC, FNAL : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

1. Generators: SWGuideCMSDataAnalysisSchool2016GeneratorExerciseatFNAL

Description: An introduction to the physics of datasets and an exploration of Monte Carlo truth information.


Responsible at CMSDAS@LPC, FNAL : Si Xie ( CALTECH), Altan Cakir ( ITU), Kevin Pedro (Fermilab)

2. Pileup and MET : SWGuideCMSDataAnalysisSchoolLPC2016PileupAndMETExercise

Description: The goal of this exercise is to familiarize you with the tools available for measuring and mitigating pileup in an event. To this end, we will be discussing; where pileup comes from, how we simulate pileup for MC, and how it is measured in data. There are several variables that are strongly correlated to pileup and we will discuss the merits of each as well as how to access them. Additionally, we will be covering MET and the corrections made to MET in response to pileup corrections made to jets. Finally, this course will cover some new tools for calculating muon isolation and how this relates to pileup.

For each of these topics we will be discussing run 2 methods, which can be implemented on the MiniAOD data tier.


Responsible at CMSDAS@LPC, FNAL : Alexx Perloff (Texas A&M University), Nhan Tran (FNAL), Justin Pilot (University of California Davis) , Ashley Parker (SUNY Buffalo)

3. Electrons+Photons: SWGuideCMSDataAnalysisSchoolLPC2016ElectronsAndPhotons

Description: This short exercise goes through the steps of accessing and displaying kinematic variables for electrons and photons, using ID criteria to improve signal efficiency and background rejection, and creating and improving a dielectron invariant mass distribution.


Responsible at CMSDAS@LPC, FNAL : Rami Kamalieddin (NEBRASKA), Lovedeep K. Saini (KSU), Marc Weinberg (Florida State University)

4. Muons: SWGuideCMSDataAnalysisSchoolLPC2016Muons

Description: This exercise has the goal to familiarize you with muon identification and isolation in CMS. It starts with a presentation describing the muon reconstruction and ID in CMS and then some interactive exercises are being performed to demonstrate the details. The first interactive part of this exercise is to learn to use muons, muon discriminators and muon isolation and to be able to choose the best recipe depending on the analysis you want to use. The exercise will conclude with a discussion on the measurement of muon efficiencies and the application of muon calibration and scale corrections.


Responsible at CMSDAS@LPC, FNAL : Norbert Neumeister (Purdue University), Marco De Mattia (Texas A&M University), Nicola De Filippis (INFN Bari)

5. Tracking and Primary Vertices: SWGuideCMSDataAnalysisSchoolLPC2016TrackingAndVertexingExercise

Description: We will present an introduction to using tracks for analyses in the era of large pile-up (many primary vertices). Our exercises will all use real data and will familiarize you with the following techniques:

  • Extracting basic track parameters and reconstructing invariant masses from tracks in CMSSW. Tracks are the closest detector entity to the four-vectors of particles: the momentum of a track is nearly the momentum of the charged particle itself.
  • Extracting basic parameters of primary vertices. In this high-luminosity era, it is not uncommon for a single event to contain as many as ten to twenty independent collisions. For most analyses, only one is relevant, and it can usually be identified by its tracks.
  • Reconstructing secondary vertices to measure the Kshort mass


Responsible at CMSDAS@LPC, FNAL : Matthew Walker (Rutgers University), Jamie Antonelli (Ohio State University), Marco De Mattia (Texas A&M University), Giuseppe Cerati (University of California San Diego)

6. Visualization and Event Display: SWGuideCMSDataAnalysisSchoolLPC2016EventDisplayExercise

Description: In this short exercise you learn how to explore interesting events using a physics event display. Event displays help understand the physics processes and particle interactions you are investigating. They are also a great tool to scan the detector for performance issues. Physics objects, like Electrons, Muons, Jets and more, can be visualized along with the various sub-detector systems. We will use Fireworks, the CMS event-display, to scan through a number of 13TeV Monte-Carlo- and data-events and in order to produce some nice snapshots of the CMS detector in action.


Responsible at CMSDAS@LPC, FNAL : Heiner Tholen (University of Hamburg), Sinan Sagir (Brown University)

7. Trigger: SWGuideCMSDataAnalysisSchoolLPC2016TriggerExercise

Description: This exercise is intended to familiarize you with finding trigger related information relevant for your physics analysis, accessing them from within your analyzer in view of using them to estimate the trigger efficiency for any given analysis. In the final discussion we will use what learned in the exercise and try to elaborate about data driven methods to estimate the trigger efficiency, or to provide correction factors to the MC one, and systematic effects associated to the possible methods chosen to compute such trigger efficiency.


Responsible at CMSDAS@LPC, FNAL : Dominick Olivito (University of California San Diego), Lesya Shchutska (University of Florida), Matthew Carver (University of Florida)

8. B Tagging: SWGuideCMSDataAnalysisSchoolLPC2016BTaggingExercise

Description: This exercise provides a pedagogical introduction to b tagging and a hands-on experience accessing and plotting b-tag discriminators for various b tagging algorithms. Furthermore, students will learn how to plot the b-tagging efficiency and the mistag rate as a function of jet kinematics as well as how to compare the performance of different b tagging algorithms. Finally, the exercise will provide a comprehensive reference to more advanced topics and additional resources.


Responsible at CMSDAS@LPC, FNAL : Ivan Marchesini (Hamburg University), Ben Kreis (Fermi National Accelerator Lab.), Caterina Vernieri (Fermi National Accelerator Lab.)

9. Statistics: SWGuideCMSDataAnalysisSchoolLPC2016Statistics

Description: In these exercises you will learn how to use the statistical modeling toolkit RooFit and the associated statistical analysis toolkit RooStats to perform some standard, but very widely used, statistical procedures, including fitting, estimating (i.e., measuring) parameters, computing intervals, and setting limits. You will also learn how to use the tool combine, which, as its name suggests, can be used to perform statistical analyses of multiple results.


Responsible at CMSDAS@LPC, FNAL : Edward Laird (Brown University), Jacob Linacre (Fermi National Accelerator Lab.), Harrison Prosper (Florida State University)

10. Taus: SWGuideCMSDataAnalysisSchoolLPC2016TauExercise

Description: Jet to tau fake rate measurement in data and MC simulation using 13 TeV data samples. These measurements will include both measurement of the gluon jet fake rate (usually from multi-jet QCD events) and quark jet fake rate (from the jets which are accompanied by W boson, i.e. in W+Jet events). Lesson to be learned: Participants will learn how a tau candidate is reconstructed in CMS and what are the main properties of the hadronic tau decays. They will learn about possibility of the jet to be mis-identified as a tau candidate and measure this probability (jet-to-tau fake rate) and can measure this rate in terms of the several quantities like tau-Pt, tau-Eta,tau-Phi and number of reconstructed vertices in the events. Additionally they will be able to compare the tau fake for gluon and quark jet and compare the fake rate in data and MC.


Responsible at CMSDAS@LPC, FNAL : Abdollah Mohammadi (Kansas State University), Laura Dodd (University of Wisconsin Madison), Adrian Perieanu (Hamburg University)

11. Jets: SWGuideCMSDataAnalysisSchoolJetAnalysisLPC2016

Description:
  • A 101 on how to access jets in the CMS framework without assuming prior knowledge of jet analysis.
  • Make you familiar with basic jet types and algorithms and how to use them in your analysis.
  • Illustrate each exercise using real life example scripts.
  • Give you a comprehensive reference to more advanced workbook examples, additional resources, and pedagogical documentation in one place.


Responsible at CMSDAS@LPC, FNAL : Salvatore Rappoccio (SUNY Buffalo), Robin Erbacher (University of California Davis), Julie Hogan (Brown University), Maral Alyari (SUNY Buffalo), Charles Harrington (SUNY Buffalo), Emmanuele Usai (University of Hamburg), Andrew Whitbeck (Fermi National Accelerator Lab.)

12. Writers PUB: SWGuideCMSDataAnalysisSchoolLPC2016WritersPubExercise

Long Exercises at LPC

This is a template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT TO DESCRIBE YOUR LONG EXERCISE/FACILITATORS IN THIS SECTION BELOW) - NameOfLongExercises : twiki url

Description: CUT PASTE DESCRIPTION HERE

Recommended Pre-requisite Short Exercises: For example Electrons, b-tagging etc.


Responsible at CMSDAS@LPC, FNAL : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

1. Top quark mass measurement from b jets and simple two-body kinematics : SWGuideCMSDataAnalysisSchool2016TopExercise

Description: Students will learn how to measure the top quark mass from the latest 13 TeV data using b-jets and simple two-body kinematics. The analysis will include proper treatment of backgrounds and systematic uncertainties involved in this analysis with collision data. The analysis performed at this school is likely to be the first top quark measurement ever at 13 TeV, and students are invited to continue this analysis all the way to a public result in the Spring of 2016, as part of the larger CMS analysis team. The kiematical details of the this method and idea are given at the following link http://arxiv.org/abs/1209.0772

Recommended Pre-requisite Short Exercises: Statistics, Jets, b-tagging


Responsible at CMSDAS@LPC, FNAL : Elvire Bouvier (Lyon), Jacob Linacre (FNAL), Maral Alyari (FNAL), Ashley Parker (SUNY Buffalo), Chad Harrington (FNAL), Rami Kamalieddin (Nebraska)

2. ZprimeToDiLeptons : SWGuideCMSDataAnalysisSchoolLPC2016ZprimeDiLeptons

Description: The Zprime to dilepton exercise focusses on the search for heavy resonances decaying to muon pairs using data collected in 2015 at 13 TeV, with the CMS detector. The students wll become familiar with the analysis strategy optimized for the Run2 LHC data. Data-driven methods used to determine the detector resolution and background composition, as well as predictions derived from simulation, will be described and tested. Results will discussed in term of the final invariant mass plots and with the exclusion limits on the Zprime mass for several new Physics Models.

Recommended Pre-requisite Short Exercises: Muon, Generator, Statistics


Responsible at CMSDAS@LPC, FNAL : Nicola De Filippis (INFN Bari), Norbert Neumeister (Purdue U.), Ivan Marchesini (University of Hamburg), John Smith (UCDavis)

3. Dark Matter mono-photons : https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2016MonophotonDM

Description: This analysis describes search for new physics in photon + MET final state. This exercise is based mainly on data collected for Run2. Students will perform the efficiency measurements for the selection, background estimation, control region studies to cross check the major background estimates (Zgamma and Wgamma) and then set the limits on dark matter and extra dimension models.

Recommended Pre-requisite Short Exercises: Electrons/Photons, Statistics, MET


Responsible at CMSDAS@LPC, FNAL : Bhawna Gomber (University of Wisconsin-Madison), Lovedeep K. Saini (KSU), Marc Weinberg (Florida State University)

4. SUSY hadronic: multijets+MET : https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolLPC2016SUSYHadronic

Description: This analysis is a generic search for new physics. It is motivated by models of R-parity conserving Supersymmetry (SUSY) that generally assume the existence of additional elementary particles. The SUSY signature we are searching for in the detector consists of several jets with high pT, large missing transverse momentum due to the LSPs, and no leptons. The data-based determination of the SM backgrounds is one of the key features of this analysis. This exercise will follow the recent analysis of the 2.1 fb-1 of data collected in 2015 at a center-of-mass energy of 13 TeV. The focus of the exercise will be on the background determination methods which are the heart of this analysis as well as the interpretation of the search results for SUSY models.

Recommended Pre-requisite Short Exercises: Jets, PileUp /MET and statistics short exercises


Responsible at CMSDAS@LPC, FNAL : Kenichi Hatakeyama (Baylor University), Altan Cakir (Istanbul Technical University), Andrew Whitbeck (Fermilab), Nhan Tran (Fermilab), Markus Stoye (CERN), Kevin Pedro (Fermilab), Jack Bradmiller-Feld (UCSB), Hongxuan Liu (Baylor University), Bibhuprasad Mahakud (TIFR), James Hirschauer (Fermilab)

5. B2G Boosted Z'->ttbar semileptonic : SWGuideCMSDataAnalysisSchool2016B2GTTbar

Description: The objective of this exercise is to perform a search for top-antitop resonances in the 13 TeV LHC Run 2. Students will perform an efficiency measurement for signal MC, optimize event selections to maximize S/sqrt(B), investigate top and W tagging algorithms, estimate the backgrounds, and perform a likelihood minimization to search for new physics in the top-antitop spectrum.

Recommended Pre-requisite Short Exercises: Statistics, Tracking/Vertexing, Electron/Photon, Muon, Jets, b-tagging short exercises


Responsible at CMSDAS@LPC, FNAL : Salvatore Rappoccio (SUNY Buffalo), Alexander Schmidt (University of Hamburg), Heiner Tholen (University of Hamburg), Justin Pilot (University of California Davis), James Dolen (SUNY Buffalo), Emanuele Usai (University of Hamburg), Dominik Nowatschin (University of Hamburg)

6. Higgs->bb : https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolLPC2016Higgsbbbar2016Exercise

Description: The observation of a Higgs-like boson with a mass of 125 GeV was an extraordinary event for the CMS experiment and for particle physics as a whole. We must now characterize the properties of this new boson and confirm or reject the hypothesis that it is the Standard Model (SM) Higgs boson. One fundamental property to study is the new boson's coupling to fermions. Do fermions receive mass through the SM Higgs mechanism? We will focus on the search for the H->bb decay, which has the largest branching ratio (~60% at 125 GeV). The search is conducted in the final state where the Higgs boson (H) is produced in association with a vector boson (V = W or Z). This is commonly referred to as the VH(bb) analysis. In this exercise, we will start with an overview of the differences between signal and background processes and the corresponding discriminating variables. We will define our selection criteria in order to maximize our sensitivity to signal and use control regions in the data to estimate the backgrounds. We will perform a fit using the dijet invariant mass as the discriminant to observe an excess or set upper limits on the Higgs boson production cross section.

Recommended Pre-requisite Short Exercises:


Responsible at CMSDAS@LPC, FNAL : Ben Kreis (Fermi National Accelerator Lab.), Caterina Vernieri (Fermi National Accelerator Lab.), Julie Hogan (Brown University), John Stupak (Purdue University Calumet), Alexx Perloff (Texas A&M University)

7. Exotica Displaced Vertices : SWGuideCMSDataAnalysisSchoolLPC2016ExoticaDisplacedVertices

Description: In this exercise, you will perform a search for a heavy resonance (M) decaying to two long-lived massive neutral particles (m), which then decay to leptons (electrons and muons). The full chain is M -> m m -> 4 leptons, with both pairs of leptons being displaced from the primary vertex. This highly exotic signature would be clear evidence of physics beyond the Standard Model, and is motivated by hidden valley scenarios.

You will perform the complete analysis with full statistics. The exercise will cover MC simulation, kinematics, definition of acceptance, efficiencies, and will use the RooStats tools to set upper limits (unless we see something new). We will discuss and explore sources of systematic uncertainty and how to estimate them using data-driven methods wherever possible. Participants will have the opportunity to be the first to explore this parameter space in the Run 2 data, since there is no active analysis targeting this final state.

Recommended Pre-requisite Short Exercises: Tracking + Primary Vertices, Muons, Electrons + Photons


Responsible at CMSDAS@LPC, FNAL : Jamie Antonelli (Ohio State), Juliette Alimena (Ohio State), Marco Di Mattia (Texas A&M), Chris Hill (Ohio State)

8. Z to tau tau cross-section measurement at 13TeV : SWGuideCMSDataAnalysisSchoolLPC2016LongExerciseTau

Description: The aim is to measure Z to tau tau cross-section @ 13 TeV. What we plan to do is select Z -> tau tau -> mu tau events both in data (single muon dataset) and MC (DYJets as signal and QCD, WJet and TTbar, and possibily diboson as backgrounds) Subtract all background from data. The difference between data and background will be considered as Z to tau tau. All backgrounds are estimated from MC, except QCD, which is computed from data using the mu-tau same-sign control region.

Recommended Pre-requisite Short Exercises: Tau


Responsible at CMSDAS@LPC, FNAL : Abdollah Mohammadi (Kansas State University), Laura Dodd (University of Wisconsin Madison), Adrian Perieanu (Hamburg University), Zaixing Mao (Brown University), Edward Laird (Brown University)

9. SUSY Same-sign dilepton searches : SWGuideCMSDataAnalysisSchool2016SameSignDileptons

Description: The goal of this exercise is to perform a search for new physics with same-sign dilepton final state. This final state could manifest itself in a range of scenarios: supersymmetry particles decay, Majorana neutrinos, top quark partners with charge 5/3 and so on. The example signal to be used in this exercise is gluino-pair production with each gluino decaying to a top-quark pair and a stable neutralino. But the overall analysis design is kept general in order to be open to other possibilities. The main focus of the exercise is on how to estimate the standard model backgrounds in this search, and how to design a search strategy in order to maximize sensitivity to possible new physics scenarios.

Recommended Pre-requisite Short Exercises: Electrons/Photons, Muons, Statistics


Responsible at CMSDAS@LPC, FNAL : Lesya Shchutska (University of Florida), Sadia Khalil (Kansas State University), Anthony Barker (Purdue University), Matthew Walker (Rutgers University), Matthew Carver (University of Florida), Giuseppe Cerati (University of California San Diego)

CMSDAS 2015 (LPC, Bari)

Pre-Workshop Exercises

Short Exercises at LPC

This is a template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT TO DESCRIBE YOUR SHORT EXERCISE/FACILITATORS IN THIS SECTION BELOW) - NameOfShortExercise : twiki url

Description: CUT PASTE DESCRIPTION HERE


Responsible at CMSDAS@LPC, FNAL : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

1. Muons: SWGuideCMSDataAnalysisSchoolMuonsLPC

Description: This exercise has the goal to familiarize you with muon identification and isolation in CMS. It starts with a presentation describing the muon reconstruction and ID in CMS and then some interactive exercises are being performed to demonstrate the details. The first interactive part of this exercise is to learn to use muons , muon discriminators and muon isolation and to be able to choose the best recipe depending on the analysis you want to use. The exercise will conclude with a discussion on the measurement of muon efficiencies and the application of muon calibration and scale corrections.


Responsible at CMSDAS@LPC, FNAL: F. Golf (UCSB), N. Neumeister (Purdue), Reddy Pratap Gandrajula (University of Iowa), Amandeep K. Kalsi (PU)

2. Electrons+Photons: https://twiki.cern.ch/twiki/bin/viewauth/CMS/SWGuideCMSDataAnalysisSchoolShortElectronExerciseLPC2015

Description: This short exercise goes through the steps of accessing and displaying kinematic variables for electrons and photons, using ID criteria to improve signal efficiency and background rejection, and creating and improving a dielectron invariant mass distribution.


Responsible at CMSDAS@LPC, FNAL: B. Gomber (UW), L. K. Saini (KSU), S. Toda (KSU), S. Khalil (KSU),

3. Tracking and Primary Vertices https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolTracker2015Exercise


Description: We will present an introduction to using tracks for analyses in the era of large pile-up (many primary vertices). Our exercises will all use real data and will familiarize you with the following techniques:

  • Extracting basic track parameters and reconstructing invariant masses from tracks in CMSSW. Tracks are the closest detector entity to the four-vectors of particles: the momentum of a track is nearly the momentum of the charged particle itself.
  • Extracting basic parameters of primary vertices. In this high-luminosity era, it is not uncommon for a single event to contain as many as ten to twenty independent collisions. For most analyses, only one is relevant, and it can usually be identified by its tracks.
  • Reconstructing secondary vertices to measure the Kshort mass


Responsible at CMSDAS@LPC, FNAL: C. Vernieri (FNAL), G. Cerati (UCSD), G. Benelli (CERN)

4. Jets : SWGuideCMSDataAnalysisSchoolJetAnalysisLPC2015

Description:

  • A 101 on how to access jets in the CMS framework without assuming prior knowledge of jet analysis.
  • Make you familiar with basic jet types and algorithms and how to use them in your analysis.
  • Illustrate each exercise using real life example scripts.
  • Give you a comprehensive reference to more advanced workbook examples, additional resources, and pedagogical documentation in one place.


Responsible at CMSDAS@LPC, FNAL : Jim Dolen(Buffalo), Justin Pilot (UCDavis), Sal Rappoccio (Buffalo), Nhan Tran(FNAL), Dan Duggan (Rutgers)

5. b tagging: SWGuideCMSDataAnalysisSchool2015BTaggingExercise

Description: This exercise provides a pedagogical introduction to b tagging and a hands-on experience accessing and plotting b-tag discriminators for various b tagging algorithms. Furthermore, students will learn how to plot the b-tagging efficiency and the mistag rate as a function of jet kinematics as well as how to compare the performance of different b tagging algorithms. Finally, the exercise will provide a comprehensive reference to more advanced topics and additional resources.


Responsible at CMSDAS@LPC, FNAL : Dinko Ferencek (Rutgers), Meenakshi Narain (Brown), Francisco Yumiceva (Florida Tech), Yiwen Wen (Peking Univ.)

6. Roostat : SWGuideCMSDataAnalysisSchoolStatistics2015LPC

Description: Introduction to limit setting and statistics tools. Try out using the combine tool and RooStats to produce limits using various statistical techniques and including systematic errors.


Responsible at CMSDAS LPC@FNAL : Phil Dudero ( TTU), Greg Landsberg (Brown), Jean-Roch Vlimant (CalTech)

7. Triggers : SWGuideCMSDataAnalysisSchool2015HLTExerciseFNAL

Description: This exercise is intended to familiarize you with finding trigger related information relevant for your physics analysis, accessing them from within your analyzer in view of using them to estimate the trigger efficiency for any given analysis. In the final discussion we will use what learned in the exercise and try to elaborate about data driven methods to estimate the trigger efficiency, or to provide correction factors to the MC one; systematic effects associated to the possible methods chosen to compute such trigger efficiency; and a few strategies to keep in mind when proposing a possible new trigger for a future analysis.


Responsible at CMSDAS LPC@FNAL : Aram Avetisyan, Tulika Bose, Dylan Rankin, Clint Richardson (Boston U.), Souvik Das (U. of Florida)

8. Pileup : SWGuideCMSDataAnalysisSchoolPileupExerciseLPC2015

Description:

  • An introduction on what pileup is and how to quantify it
  • We will cover how to access pileup information and what each variable represents
  • We will also cover pileup corrections to jets and muon isolation variables
    • For each of these topics will cover the run 1 methods as well as the state of the art run 2 methods
  • Each section/topic will use example scripts to illustrate the concepts
  • Within the code and on the twiki there will be links to additional resources


Responsible at CMSDAS@LPC, FNAL : Nhan Tran (FNAL), Mike Hildreth (Notre Dame), Alexx Perloff (Texas A&M University)

9. Generators: SWGuideCMSDataAnalysisSchool2015GeneratorExerciseatFNAL

Description: An introduction to the physics of datasets and an exploration of Monte Carlo truth information.


Responsible at CMSDAS@LPC, FNAL : Stephen Mrenna ( FNAL), Altan Cakir ( DESY)

10. Writers PUB: https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolWriersExerciseLPC2015

Description:


Responsible at CMSDAS@LPC, FNAL : Dave Cutts (Brown), Lenny Spiegel (FNAL), Brajesh Choudhary (Delhi), Greg Landsberg (Brown), Tulika Bose (BU).

11. Visualisation : SWGuideCMSDataAnalysisSchool2015EventScanningExercise

Description: The goal of this short exercise is to enable students to "scan" interesting events using a physics event display. Event displays help to understand how the detector works by visualizing physics processes and particle interactions with various sub-detectors. The tool to be used is Fireworks, which is the CMS event-display project and cmsShow is the official name of the executable. This tool has been used in early cosmic ray events (CRAFT) and then in early running at 0.45 and 2.36 TeV. From the period April – October 2010, the LHC luminosity was doubling every ~2 weeks. During this period, an "Exotica Scan" was set up, which selected the ~ 30 most bizarre events each day and asked a small group of "scanners" to try to understand them. The goal was to identify problems and issues with the CMS detector. In this exercise, we will use Fireworks to scan Monte Carlo event to understand how an event will looks like in the detector.


Responsible at CMSDAS@LPC and : Zhenbin Wu (UIC), Giuseppe Cerati (UCSD) Sinan Sagir (Brown University)

Long Exercises at LPC

This is a template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT TO DESCRIBE YOUR LONG EXERCISE/FACILITATORS IN THIS SECTION BELOW) - NameOfLongExercises : twiki url

Description: CUT PASTE DESCRIPTION HERE

Recommended Pre-requisite Short Exercises: For example Electrons, b-tagging etc.


Responsible at CMSDAS@LPC, FNAL : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

1. SUSY hadronic (multijets+MET) https://twiki.cern.ch/twiki/bin/viewauth/CMS/SWGuideCMSDataAnalysisSchool2015SUSYJetsPlusMHTLPC

Description: This analysis is a generic search for new physics. It is motivated by models of R-parity conserving Supersymmetry (SUSY) that generally assume the existence of additional elementary particles. The SUSY signature we are looking for in the detector consists of several jets with high pT, large missing transverse momentum due to the LSPs, and no leptons. The data-based determination of the SM backgrounds is one of the key features of this analysis. This exercise will follow the recently published analysis of the 19.5/fb of data collected in 2012 at a center-of-mass energy of 8 TeV. The focus of the exercise will be on using 13 TeV MC samples to understand (some of) the background determination methods which are the heart of this analysis.

Prerequisites: Students should benefit from jets, MET or RooStat short exercises


Responsible at CMSDAS@LPC : H. Liu (Baylor U.), F. Lacroix (UCR), A. Cakir (DESY), A. Whitbeck (Fermilab), S. Malik (U. of Puerto Rico)

2. Top Quark: ttbar cross section: https://twiki.cern.ch/twiki/bin/viewauth/CMS/SWGuideCMSDataAnalysisSchoolTopExercise

Description: Students will learn how to measure the top quark pair cross section using the 8 TeV data and extrapolate the results of the analysis to 13 TeV. The student also will learn to estimate the main backgrounds and systematic uncertainties involved in this analysis with collision data. The measurement of the top pair cross section is one of the early priority analysis in CMS that will be done with the first 1/fb of data in 2015.

Recommended Short Exercises: b-tagging, muons, Roostat, jets


Responsible at CMSDAS@LPC, FNAL : Francisco Yumiceva (Florida Institute of Technology), Freya Bleckman (Brussels), Gabrielle Benelli (CERN), Sadia Khalil (KSU)

3. SUSY: Razor search with jet substructure: https://twiki.cern.ch/twiki/bin/viewauth/CMS/SWGuideCMSDataAnalysisSchool2015RazorExerciseFNAL

Description: Student will learn how to use jet substructure in a search for new physics. The specific case of boosted W bosons from stop decays will be considered. The variables used for boosted-W tagging, accessed from miniAOD, will be ntuplized and used for event selection in a SUSY-sensitive signal region. The signal region will be defined using the razor variables R and MR. The student will learn about the variable definition, the properties of signal and background distribution and how the correlation helps to enhance the search sensitivity. The effect of W tagging on the event kinematic shape will be illustrated. The final target of the exercise consists in the definition of a signal region, for which an expected sensitivity to gluino and stop production at 13 TeV will be estimated.

Recommended Pre-requisite Short Exercises: jets and MET short exercises


Responsible at CMSDAS@LPC, FNAL : SEZEN SEKMEN (), MAURIZIO PIERINI (CALTECH/LPC), JUSTIN PILOT (), JAVIER DUARTE (CALTECH)

4. B2G : ttbar resonances with jet substructure : SWGuideCMSDataAnalysisSchool2015B2GTTbar

Description: The objective of this exercise is to perform a search for top-antitop resonances in the 13 TeV LHC Run 2. This exercise is MC-only, in preparation for the upcoming Run 2. Students will perform an efficiency measurement for signal MC, optimize event selections to maximize S/sqrt(B), investigate top and W tagging algorithms, estimate the backgrounds, and perform a likelihood minimization to search for new physics in the top-antitop spectrum.

Recommended Pre-requisite Short Exercises:


Responsible at CMSDAS@LPC, FNAL : Sadia Khalil (Kansas), James Dolen (Buffalo), Salvatore Rappoccio ( Buffalo), Cecilia Gerber (UIC), Nhan Tran (FNAL), Paul Turner (UIC), Daniel Sandoval (UIC)

5. EXO: Dijet resonances (no substructure): SWGuideCMSDataAnalysisSchool2015EXODijetResonances

Description: The purpose of this long exercise is to build a foundation for performing analyses at a hadron collider detector. We will do this by focusing on the ongoing search at CMS for dijet resonances. The final goal is to teach students how to do a simple search for a dijet resonance signal. The basic physics of dijet resonance signals and QCD background will be taught, and the analysis techniques used to analyze the data will be explored. We will use the technical experience gained in the pre-exercises and short-exercises as a basis for the long exercise. But unlike the earlier exercises, we will emphasize analysis techniques rather than strictly the software machinery. Of course, we will continue to develop and hone the machinery as we go along, but we will treat CMS software and ROOT as a means rather than an end. The course design is intended for starting graduate students with little or no prior experience in high energy physics analysis techniques.


Responsible at CMSDAS@LPC, FNAL : Dinko Ferencek (Rutgers), Artur Apresyan (Caltech), Dan Duggan (Rutgers), Jean-Roch Vlimant (Caltech), Robert Harris (FNAL)

6. B2G: Same-sign dilepton searches : SWGuideCMSDataAnalysisSchool2015B2GSameSignDileptons

Description: This exercise illustrates how to perform a search exotic signatures which decay to same-sign dileptons. The signal used is a top partner with charge 5/3 (the T5/3), but the techniques used to estimate the backgrounds due to fake and non-prompt leptons and prompt leptons with an incorrectly measured charge are common to many same-sign analyses.

Recommended Pre-requisite Short Exercises: Electrons, muons


Responsible at CMSDAS@LPC, FNAL : Aram Avetisyan, Tulika Bose, Dylan Rankin, Clint Richardson (Boston U.), Meenakshi Narain (Brown U.), Sam Bein (Florida State U.)

7. DarkMatter - Monophoton : SWGuideCMSDataAnalysisSchool2015MonoPhotonDM

Description: This analysis describes search for new physics in photon + MET final state. This exercise is based mainly on MC for Run2. Students will perform the efficiency measurements for the selection, background estimation, control region studies to cross check the major background estimates (Zgamma and Wgamma) and then set the limits on dark matter models.

Recommended Pre-requisite Short Exercises: photons, MET, statistics


Responsible at CMSDAS@LPC, FNAL : Bhawana Gomber (UW), Marc Weinberg (FSU), Andrew Askew (FSU), Arka Santra (FSU)

8. Higgs to bbar https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolHiggsbbbar2015Exercise

Description: The observation of a Higgs-like boson in 2012 with a mass of 126 GeV was an extraordinary event for the CMS experiment and for particle physics. The next goal of the experiment is to characterize the properties of this new boson, and confirm or reject the hypothesis that it is the Standard Model (SM) Higgs. One fundamental property to study is the new boson's coupling to fermions. We will focus on the search for the H->bb decay, which has the largest branching ratio (~60% at 125 GeV). This search is conducted in the final state where H is produced in association with a vector boson (V = W or Z). This is commonly referred to as the VH(bb) analysis. In this Exercise, we will start with an overview of the differences between signal and background processes, and the corresponding discriminating variables. We will define our selection criteria in order to maximize our sensitivity to signal in the data. We will perform a fit using the dijet invariant mass as the discriminant to set upper limits on the Higgs production cross section (or observe an excess compatible with the SM Higgs). If time permits, we will also perform an analysis of the diboson process VZ(bb) that provides a crucial calibration for the H ->bb search. This Exercise is based on the LHCP 2013 m(jj) analysis in the highest boost regions with some simplification.

Recommended Pre-requisite Short Exercises: b-tagging, roostat, jets


Responsible at CMSDAS LPC@FNAL : Caterina Vernieri (FNAL) Souvik Das (Florida) Nhan Tran (FNAL) John Stupak (Purdue)

9. Exotica displaced v. https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2012ExoticaDisplacedVertices

Description: In this exercise, you will perform a search for a heavy resonance (M) decaying to two long-lived massive neutral particles (m), which then decay to leptons (electrons and muons). The full chain is M -> m m -> 4 leptons, with both pairs of leptons being displaced from the primary vertex. This highly exotic signature would be clear evidence of physics beyond the Standard Model, and is motivated by hidden valley scenarios.

You will perform the complete analysis with full statistics. The exercise will cover MC simulation, kinematics, definition of acceptance, efficiencies, and will use the RooStats tools to set upper limits (unless we see something new). We will discuss and explore sources of systematic uncertainty and how to estimate them using data-driven methods wherever possible.

The exercise will be based on 2011 and 2012 data. The increased energy and statistics will lead to an increased sensitivity and provide a first look beyond the existing limits which were derived on 2011 data.

Recommended Pre-requisite Short Exercises: tracking, vertexing, muons, statistics.


Responsible at CMSDAS@LPC, FNAL : Zhen Hu (Fermilab), Jamie Antonelli (The Ohio State University), Reddy Pratap Gandrajula (University of Iowa)

Short Exercises at Bari

This is a template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT TO DESCRIBE YOUR SHORT EXERCISE/FACILITATORS IN THIS SECTION BELOW) - NameOfShortExercise : twiki url

Description: CUT PASTE DESCRIPTION HERE


Responsible at CMSDAS@INFN, Bari : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

1. MET: CMSPublic.SWGuideCMSDAS2015METBari

Description: This short exercise introduces the basics of MET reconstruction and corrections and teaches how to use MET software in CMSSW.


Responsible at CMSDAS@INFN Bari: T. Sakuma (Bristol U.)

2. Electrons : SWGuideCMSDataAnalysisSchool2015ElectronExercise (at Bari)

Description: This short exercise goes through the steps of accessing and displaying kinematic variables for electrons and photons, using ID criteria to improve signal efficiency and background rejection, and creating and improving a dielectron invariant mass distribution.


Responsible at CMSDAS@INFN, Bari : S. Baffioni (LLR)

3. Muons: SWGuideCMSDataAnalysisSchoolMuonsBari

Description: This exercise has the goal to familiarize you with muon identification and isolation in CMS. It starts with a presentation describing the muon reconstruction and ID in CMS and then some interactive exercises are being performed to demonstrate the details. The first interactive part of this exercise is to learn to use muons , muon discriminators and muon isolation and to be able to choose the best recipe depending on the analysis you want to use. Then there is a short exercise on the muon calibration and scale corrections that demonstrates how the muon scale is improved..


Responsible at CMSDAS@INFN Bari: C. Calabria/P. Verwilligen/S. Chhibra (INFN Bari), C. Battilana (CIEMAT)

4. b tagging : SWGuideCMSDataAnalysisSchool2015BTaggingExercise

Description: This exercise provides a pedagogical introduction to b tagging and a hands-on experience accessing and plotting b-tag discriminators for various b tagging algorithms. Furthermore, students will learn how to plot the b-tagging efficiency and the mistag rate as a function of jet kinematics as well as how to compare the performance of different b tagging algorithms. Finally, the exercise will provide a comprehensive reference to more advanced topics and additional resources.


Responsible at CMSDAS@INFN Bari: S. Donato (INFN Pisa), I. Marchesini (DESY)

5. Triggers : SWGuideCMSDataAnalysisSchool2015HLTExerciseFNAL SWGuideCMSDataAnalysisSchool2015HLTExerciseBari

Description: This exercise is intended to familiarize you with finding trigger related information relevant for your physics analysis, accessing them from within your analyzer in view of using them to estimate the trigger efficiency for any given analysis. In the final discussion we will use what learned in the exercise and try to elaborate about data driven methods to estimate the trigger efficiency, or to provide correction factors to the MC one; systematic effects associated to the possible methods chosen to compute such trigger efficiency; and a few strategies to keep in mind when proposing a possible new trigger for a future analysis.


Responsible at CMSDAS@INFN Bari: M. De Gruttola (CERN), L. Guiducci (INFN Bologna)

6. Roostat/RooFit : SWGuideCMSDataAnalysisSchoolStatistics2015LPC, SWGuideCMSDataAnalysisSchoolStatistics2015Bari (prev. edition: SWGuideCMSDataAnalysisSchoolStatistics2014)

Description: Introduction to limit setting and statistics tools. Try out using the combine tool and RooStats to produce limits using various statistical techniques and including systematic errors.


Responsible at CMSDAS@INFN Bari: L. Lista (INFN Napoli), M. Pelliccioni (INFN Torino)

7. Generators : SWGuideCMSDataAnalysisSchool2015GeneratorExerciseatBari

Description: An introduction to the usage of Monte Carlo generators in CMSSW


Responsible at CMSDAS@INFN Bari: J. Smith (UCDavis), P. Rebello Teles (CBPF Brasil), V. Ciulli (Università e INFN Firenze)

8. MVA techniques : SWGuideCMSDataAnalysisSchool2015MVAExerciseatBari

Description: The exercise will give an introduction to the most common multivariate analysis (MVA) methods. MVA methods span a broad spectrum of analysis tasks. However, for concreteness, the examples illustrate the problem of signal/background discrimination using two 2-D problems.


Responsible at CMSDAS@INFN Bari: H. Prosper (FSU), C. Caputo and S. Chhibra (INFN Bari), G. Singh (Chulalongkorn U.)

9. Tracking and Primary Vertices https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolTracker2015Exercise

Description: We will present an introduction to using tracks for analyses in the era of large pile-up (many primary vertices). Our exercises will all use real data and will familiarize you with the following techniques:

  • Extracting basic track parameters and reconstructing invariant masses from tracks in CMSSW. Tracks are the closest detector entity to the four-vectors of particles: the momentum of a track is nearly the momentum of the charged particle itself.
  • Extracting basic parameters of primary vertices. In this high-luminosity era, it is not uncommon for a single event to contain as many as ten to twenty independent collisions. For most analyses, only one is relevant, and it can usually be identified by its tracks.
  • Reconstructing secondary vertices to measure the Kshort mass


Responsible at CMSDAS@INFN Bari: G. Benelli (Texas U.)

Long Exercises at Bari

This is a template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT TO DESCRIBE YOUR LONG EXERCISE/FACILITATORS IN THIS SECTION BELOW) - NameOfLongExercises : twiki url

Description: CUT PASTE DESCRIPTION HERE


Responsible at CMSDAS@INFN, Bari : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

1. Zprime SWGuideCMSDataAnalysisSchool2015DimuonExercise

Description: This extended exercise is intended to familiarize you with two dimuon analysis efforts in CMS. The Drell-Yan to dimuon spectrum is a fruitful signature for discovering new physics, and we seek to measure the Drell-Yan spectrum to high precision in CMS. Many new models of physics predict a new narrow resonance, here referred to generically as Z', that decays into a pair of oppositely-charged muons. We look for a bump from Z' in the dimuon invariant mass spectrum using a shape-based search, looking for the difference in between the exponentially falling background and the resonance peak that would show up should a Z' exist. In this exercise, you'll work on the meat of the analysis to produce the dimuon mass spectra for data and MC. In doing this exercise you can learn (among other things): how to apply some loose and tight selection criteria to select muons and make combined dimuon object collections; how to evaluate acceptance, trigger efficiency, mass resolution; how to estimate some of the backgrounds using the data itself.

Prerequisite: "Muons" short exercise.


Responsibles at CMSDAS@INFN Bari: R. Radogna ( INFN Bari), S. Elgammal ( BUE Egypt)

2. H->ZZ->4l SWGuideCMSDataAnalysisSchoolHZZ4lSearchExercise

Description: The 4 leptons final state has been one of the main channel in the discovery of the boson at 125 GeV. The importance of this channel is due to the very small background and very good mass resolution. So, as first step, we will review the analysis flow to understand how we can select a clean signal. We will analyze the kinematics of the signal and the main background (ZZ) and we will extract the expected lepton efficiency and resolution. We will study the features of the reducible background (Z+jets) and we will use control region in data to cross-check the expected number of Z+jets events in the signal region. As second step, we will apply statistical analysis to the results to compute the p-value of the observed excess in data and, if time will permit, we will extract a measurement of the mass of the discovered resonance and the discrimination between spin/parity hypotheses.

Prerequisites: Muons/Electrons and RooFit short exercise are recommended


Responsibles at CMSDAS@INFN Bari: G. Ortona (LLR), S. Chhibra and N. De Filippis ( INFN Bari), David Sperka ( University Florida)

3. Dark Matter serach via H->invisible SWGuideCMSDataAnalysisSchoolHInvisibleSearchExercise

Description: bla bla.

Prerequisites: Muons and RooStat short exercise are recommended


Responsibles at CMSDAS@INFN Bari: P. Meridiani (INFN Rome), A.M. Magnan ( IC)

4. Dark Matter search via monojet with MET SWGuideCMSDataAnalysisSchoolMonoJetSearchExercise

Description: For the monojet exercise, we present one of analyses used to search for new physics, especially dark matter searches at the collider, in the final state containing a single jet and missing transverse energy. In the exercise, you will practice on event selection, noise reduction, and standard model background estimations using 8 TeV MC and data. The dark matter particle - nucleon cross-section will be calculated at the end, to compare with direct, and indirect detection experiments. The future analysis at 13 TeV will be discussed at the end of the exercises, there are many rooms for contributions in this analysis.

Prerequisites: Jet/MET and RooStat short exercise are recommended


Responsibles at CMSDAS@INFN Bari: P. Srimanobhas (Chulalongkorn U.)

5. SUSY hadronic

Description: This analysis is a generic search for new physics. It is motivated by models of R-parity conserving Supersymmetry (SUSY) that generally assume the existence of additional elementary particles. The SUSY signature we are looking for in the detector consists of several jets with high pT, large missing transverse momentum due to the LSPs, and no leptons. The data-based determination of the SM backgrounds is one of the key features of this analysis. This exercise will follow the recently published analysis of the 19.5/fb of data collected in 2012 at a center-of-mass energy of 8 TeV. The focus of the exercise will be on understanding (some of) the data-based background determination methods which are the heart of this analysis.

Prerequisites: Muons and RooStat short exercise are recommended


Responsible at CMSDAS@INFN Bari : M. De Gruttola (CERN), S. Donato (Scuola Normale Superiore, Pisa)

6. Higgs to bbar https://twiki.cern.ch/twiki/bin/viewauth/CMS/SWGuideCMSDataAnalysisSchoolHiggsbbbar2015ExerciseBari

Description: The observation of a Higgs-boson like in 2012 with a mass of 126 GeV has been an extraordinary event for CMS and for particle physics. The next goal of the experiment is now to characterize the properties of this new boson, and confirm/exclude that is behaving as the Standard Model (SM) Higgs. One fundamental property to study is the new boson coupling to fermions. We will focus on a H->bb search which has the largest decay branching ratio (~60% at 125 GeV). In this exercise we will start with an overview of the differences between signal and background processes and corresponding discriminating variables in the channel. We will try to define our selection cuts in order to maximize the probability to observe a signal in our data. We will perform a fit to the dijet invariant mass shape to set limits on the Higgs cross section (or observe an excess compatible with the SM Higgs) and we will look into several systematic uncertainty contributions (like b-tag efficiency, jet resolution,etc.). We will perform a dedicated analysis for the di-boson process W(Z)+ Z->bb, that provides a crucial calibration for the Higgs to bbar search. The methods used in the exercises will be very similar to the ones used in the official analysis.

Prerequisites: Muons and RooStat short exercise are recommended


Responsible at CMSDAS@INFN Bari: S. Donato (INFN Pisa), M. De Gruttola (CERN)

7. ttH https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolttH2015Exercise

Description: The observation of a Higgs-boson like in 2012 with a mass of 125 GeV has been an extraordinary event for CMS and for particle physics. The next goal of the experiment is now to characterize the properties of this new boson, and confirm/exclude that is behaving as the Standard Model (SM) Higgs. One fundamental property to study is the new boson coupling to fermions. Owing to its large mass, the top quark is expected to play an important role in the electroweak symmetry breaking mechanism. A direct evidence of the Higgs boson coupling to top quarks is still lacking: indeed, the 125 GeV Higgs boson is too light to decay to top quarks. This coupling can be however probed by measuring the production cross section of processes where both a Higgs boson and top quarks are produced in the final state. We will focus on a H->bb search which has the largest decay branching ratio (~60% at 125 GeV). In this exercise we will start with an overview of the differences between signal and background processes and corresponding discriminating variables in the channel. We will try to define our selection cuts in order to maximize the probability to observe a signal in our data. The methods used in the exercises will be very similar to the ones used in the official analysis.

Prerequisites: Muons and RooStat short exercise are recommended


Responsible at CMSDAS@INFN Bari: J. Pata (ETH)

8. X->hh->bbtautau https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolXXH2015Exercise

Description: A search for a heavy Higgs (H) decaying to two 125 GeV Higgs bosons (h) using the final state of 2 taus and 2 b is presented. The exercise will focus on the results obtained, using the 19.7~fb^{-1} of pp collision data collected during 2012 at the CMS experiment with a centre of mass energy = 8 TeV. In the signal hypothesis the two light Higgs are the 125 GeV Higgs boson discovered at the LHC. In certain high MSUSY models at low tan(beta), the decay mode H->hh is enhanced. The analysis is studied in a mass range from m_H=260 GeV (the kinematic threshold for H->h(125)h(125)) and up to m_H=350 GeV, at which point the branching ratio of the light Higgs to top quarks begins to dominate. In addition to providing an m_A-tan(beta) limit in interesting MSSM scenarios, a model-independent limit on cross-section*BR for the H->hh process is included in the same mass range..

Prerequisites: Muons, btagging and MET short exercise are recommended


Responsible at CMSDAS@INFN Bari: M.T. Grippo and K. Androsov (University of Siena and INFN Pisa), F. Ricci-Tam (UCDavis)

10: Top-AntiTop https://twiki.cern.ch/twiki/bin/view/CMSPublic/SWGuideCMSDataAnalysisSchool2015TTbarLongExercise

Measurement of the top quark pair production cross section in events containing lepton+jets using b-quark identification and top-like BSM searches.

Description: The basic exercise measures the top quark pair production cross section using events containing one isolated muon, 3 or more jets, missing transverse energy, indicative of the presence of a neutrino, and using b-quark identification. This sample can also be used to search for new physics if there is interest in doing so. Students will learn about the main background sources to the top quark signal in this channel, and use a data-driven method to estimate the QCD multijet background. The top quark pair production cross section will be measured from a fraction of the 2012 data using two methods: template fit and counting. In addition, the students will learn how to search for heavy top quark resonances.

Suggested short exercises: b-tagging, muons, particle flow, RooStats (strongly suggested).

Prerequisites: Muons and RooStat short exercise are recommended


Responsible at CMSDAS@INFN Bari: A: Popov (UCL), L. Benucci (Ghent Univ.)

Short Exercises at Korea

This is a template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT TO DESCRIBE YOUR LONG EXERCISE/FACILITATORS IN THIS SECTION BELOW) - NameOfShortExercise : twiki url

Description: CUT PASTE DESCRIPTION HERE


Responsible at CMSDAS@KNU, Korea : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

1. Electron : SWGuideCMSDataAnalysisSchool2015KoreaElectronsExercise
Responsible at CMSDAS@Korea KNU: Minsuk Kim (Kyungpook Natl. Univ.), Youn Jung Roh (Korea Univ.)

2. Muon : SWGuideCMSDataAnalysisSchool2015KoreaMuonsExercise
Responsible at CMSDAS@Korea KNU: Taejeong Kim (Cheonbuk Natl. Univ.), Junghwan Goh (Sungkyunkwan Univ.)

3. Jet/MET : SWGuideCMSDataAnalysisSchool2015KoreaJetMetExercise SWGuideCMSDataAnalysisSchool2015KoreaJetExercise
Responsible at CMSDAS@Korea KNU: SangEun Lee (Kyungpook Natl. Univ.), Jason Lee (Univ. of Seoul)

4. Trigger : SWGuideCMSDataAnalysisSchool2015KoreaTriggerExercise
Responsible at CMSDAS@Korea KNU: Hwidong Yoo (Seoul Natl. Univ.), Dong Ho Moon (Chonnam Natl. Univ.)

5. b tagging : SWGuideCMSDataAnalysisSchool2015KoreaBTaggingExercise
Responsible at CMSDAS@Korea KNU: Youngdo Oh (Kyungpook Natl. Univ.), Sehwook Lee (Korea Univ.)

Description: This exercise provides a pedagogical introduction to b tagging and a hands-on experience accessing and plotting b-tag discriminators for various b tagging algorithms. Furthermore, students will learn how to plot the b-tagging efficiency and the mistag rate as a function of jet kinematics as well as how to compare the performance of different b tagging algorithms. Finally, the exercise will provide a comprehensive reference to more advanced topics and additional resources.

Long Exercises at Korea

This is a template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT TO DESCRIBE YOUR LONG EXERCISE/FACILITATORS IN THIS SECTION BELOW) - NameOfLongExercise : twiki url

Description: CUT PASTE DESCRIPTION HERE


Responsible at CMSDAS@INFN, Bari : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

1. Top Quark: ttbar cross section: https://twiki.cern.ch/twiki/bin/viewauth/CMS/SWGuideCMSDataAnalysisSchoolTopQuark

Description: Students will learn how to measure the top quark pair cross section using the 13 TeV. The student also will learn to estimate the main backgrounds and systematic uncertainties involved in this analysis with collision data. The measurement of the top pair cross section is one of the early priority analysis in CMS that is being done with the first some pb-1 of data in 2015.

Recommended Short Exercises: b-tagging, muons, Roostat, jets


Responsible at CMSDAS@Korea : Tae Jeong Kim (Chonbuk National University), Youn Roh (Korea University), Geonmo Ryu (University of Seoul), Sehwook Lee (Korea University), Jungwhan Goh (Sungkyunkwan University)

2. SUSY: Razor search with jet substructure:


Responsible at CMSDAS@Korea : Min Suk Kim (Kyungpook Natl. Univ.), Hwidong Yu (Seoul Natl. Univ.), Sezen Sekmen (Kyungpook National University)

https://twiki.cern.ch/twiki/bin/viewauth/CMS/SWGuideCMSDataAnalysisSchool2015RazorExerciseKorea

Description: Student will learn how to use jet substructure in a search for new physics. The specific case of boosted W bosons from stop decays will be considered. The variables used for boosted-W tagging, accessed from miniAOD, will be ntuplized and used for event selection in a SUSY-sensitive signal region. The signal region will be defined using the razor variables R and MR. The student will learn about the variable definition, the properties of signal and background distribution and how the correlation helps to enhance the search sensitivity. The effect of W tagging on the event kinematic shape will be illustrated. The final target of the exercise consists in the definition of a signal region, for which an expected sensitivity to gluino and stop production at 13 TeV will be estimated.

Recommended Pre-requisite Short Exercises: jets and MET short exercises

3. Zprime SWGuideCMSDataAnalysisSchool2015KoreaDimuonExercise

Description: This extended exercise is intended to familiarize you with two dimuon analysis efforts in CMS. The Drell-Yan to dimuon spectrum is a fruitful signature for discovering new physics, and we seek to measure the Drell-Yan spectrum to high precision in CMS. Many new models of physics predict a new narrow resonance, here referred to generically as Z', that decays into a pair of oppositely-charged muons. We look for a bump from Z' in the dimuon invariant mass spectrum using a shape-based search, looking for the difference in between the exponentially falling background and the resonance peak that would show up should a Z' exist. In this exercise, you'll work on the meat of the analysis to produce the dimuon mass spectra for data and MC. In doing this exercise you can learn (among other things): how to apply some loose and tight selection criteria to select muons and make combined dimuon object collections; how to evaluate acceptance, trigger efficiency, mass resolution; how to estimate some of the backgrounds using the data itself.

Prerequisite: "Muons" short exercise.


Responsibles at CMSDAS@Korea: Jason Lee ( Univ of Seoul), Youngdo Oh ( Korea)

CMSDAS 2014 (archive) (LPC, CERN)

Pre-Workshop Exercises

Short Exercises

Template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT BELOW) - NameOfShortExercises : twiki url

Description: CUT PASTE DESCRIPTION HERE


Responsible at CMSDAS LPC@FNAL : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)
Responsible at CMSDAS CERN: FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

1. Jets @FNAL SWGuideCMSDataAnalysisSchoolJetAnalysis @CERN(Jets and MET) SWGuideCMSDataAnalysisSchoolJetMetAnalysis

Description: The short exercise provides hands-on experience accessing jets, plotting basic jet quantities, and teaches the basics of jet energy corrections and their uncertainty. You will learn the various available pile-up corrections for jets. You will become familiar with basic jet types and algorithms, including jet sub-structure algorithms.


Responsible at CMSDAS LPC@FNAL : Ajay Kumar ( University of Delhi), Alexx Perloff (Texas A&M), John Stupak (Purdue-Calumet), Nhan Tran ( FNAL)
Responsible at CMSDAS CERN: Phil Harris ( CERN),Sebastien Brochet( IPN Lyon), Konstantinos Kousouris (CERN)

2. Electrons-Photons : SWGuideCMSDataAnalysisSchool2014ElectronPhotonExerciseCERN (at CERN) SWGuideCMSDataAnalysisSchool2014ElectronPhotonExercise (at LPC@FNAL)

Description: This short exercise goes through the steps of accessing and displaying kinematic variables for electrons and photons, using ID criteria to improve signal efficiency and background rejection, and creating and improving a dielectron invariant mass distribution.


Responsible at CMSDAS LPC@FNAL : Rishi Patel (Rutgers),Marc Weinberg (Florida State)
Responsible at CMSDAS CERN: Emanuele di Marco ( CERN),David Sabes (POLYTECHNIQUE)

3. Muons

Exercise at FNAL: SWGuideCMSDataAnalysisSchool2013MuonExercise

Description: This exercise is intended to familiarize you with using muons in a physics analysis. In particular, you will learn: a brief overview of muon reconstruction and identification algorithms, what muon related information is available in AOD and PAT and how to access muon specific information that could be relevant for your analysis. We will show you what typical performance you can expect for muons in CMS, regarding efficiency, acceptance, purity and momentum resolution. The exercise will also cover the basics of measuring the muon identification and trigger efficiency in data.


Responsible at CMSDAS LPC@FNAL : Hwidong Yoo (Purdue U.), Benjamin Stieger (CERN), Martijn Mulders (CERN)

Exercise at CERN: SWGuideCMSDataAnalysisSchoolMuonsCERN

Description: This exercise has the goal to familiarize you with muon identification and isolation in CMS. It starts with a presentation describing the muon reconstruction and ID in CMS and then some interactive exercises are being performed to demonstrate the details. The first interactive part of this exercise is to learn to use muons , muon discriminators and muon isolation and to be able to choose the best recipe depending on the analysis you want to use. Then there is a short exercise on the muon calibration and scale corrections that demonstrates how the muon scale is improved..


Responsible at CMSDAS CERN: Michail Bachtis (CERN), Luca Perrozzi (CERN)

4. Generators : SWGuideCMSDataAnalysisSchool2014GeneratorExerciseatFNAL SWGuideCMSDataAnalysisSchool2014GeneratorExerciseatCERN

Description: This exercise is based on a longer tutorial developed for the Hadron Collider Physics Summer School (HCPSS) which would work outside of CMSSW. The current tutorial is shorter, but works within CMSSW. For those who are interested, the notes and software guide from the HCPSS are duplicated here. This tutorial will guide you on an investigation of the similarities and differences that can be encountered when comparing predictions made with different Monte Carlo (MC) tools. In all cases, we focus on W+ and W production and decay to leptons at the LHC (a 7 TeV, proton-proton collider).


Responsible at CMSDAS LPC@FNAL : Alexey Ferapontov (Brown U.), Guillermo Breto Rangel (UC Davis), John Smith (UC Davis), Jean-Roch Vlimant (CERN)
Responsible at CMSDAS CERN: Vitaliano Ciulli (CERN/Firenze),Fabio Cossutti (Trieste)

5. Visualisation : SWGuideCMSDataAnalysisSchool2012EventScanningExercise

Description: The goal of this short exercise is to enable the student to “scan” interesting CMS events. The tool to be used is Fireworks. This tool has been used in early cosmic ray events (CRAFT) and then in early running at 0.45 and 2.36 TeV. The representative data used here is from the period April – October 2010 when the LHC luminosity was doubling every~ 2 weeks. During that period an “Exotica Scan” was set up, which selected the ~ 30 most bizarre events each day and asked a small group of “scanners” to try to understand them. The goal was to identify problems and issues with the CMS detector. The exerises of is of general use when looking at events even at 8 TeV or at higher energy in future era.


Responsible at CMSDAS@LPC and : Sudhir Malik ( Nebraska/FNAL), Joshua Swanson ( Brown University)
Responsible at CMSDAS@CERN and : Not done at CERN

6. Tracking and Primary Vertices https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolTracker2014Exercise

Description: We will present an introduction to using tracks for analyses in the era of large pile-up (many primary vertices). Our exercises will all use real data and will familiarize you with the following techniques:

  • Extracting basic track parameters and reconstructing invariant masses from tracks in CMSSW. Tracks are the closest detector entity to the four-vectors of particles: the momentum of a track is nearly the momentum of the charged particle itself.
  • Extracting basic parameters of primary vertices. In this high-luminosity era, it is not uncommon for a single event to contain as many as ten to twenty independent collisions. For most analyses, only one is relevant, and it can usually be identified by its tracks.
  • Reconstructing secondary vertices to measure the Kshort mass


Responsible at CMSDAS LPC@FNAL : Thomas Hauth (CERN), Caterina Vernieri (Sezione di Pisa (IT)), Steve Mrenna (Fermilab)
Responsible at CMSDAS CERN: Marco Rovere (CERN), Giovanni Petrucciani (CERN), Thomas Hauth (CERN)

7. B-tagging : SWGuideCMSDataAnalysisSchoolbTagExercise

Description:

We will learn about the properties and differences between primary vertices and the luminous region (also known as beam spot). Students will learn how to access the vertex collections and make plots of the main variables. Furthermore, they will learn how to identify b-flavored jets and plot the b-tag efficiency and mistagging rate as a function of jet kinematics for a given algorithm. Use the b-tagging scale factors derived from data to correct the Monte Carlo efficiencies.


Responsible at CMSDAS LPC@FNAL: Francisco Yumiceva (Florida Tech),Steven Kaplan (Rutgers)
Responsible at CMSDAS CERN: MariaRosaria D' Alfonso (CERN), Pedro Silva (CERN), Sezen Sekmen (CERN)

8. Pile-up https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2014PileupReweighting

Description: Learn the procedure for reweighting MC distributions to match the pileup in the data. Students will learn how to access the various pieces of information related to pileup in the MC events, and how to use the reweighting tools.

Responsible at CMSDAS LPC@FNAL: Mike Hildreth ( Notre Dame)

11. Triggers :

Description: This exercise is intended to familiarize you with finding trigger related information relevant for your physics analysis, accessing them from within your analyzer in view of using them to estimate the trigger efficiency for any given analysis. In the final discussion we will use what learned in the exercise and try to elaborate about data driven methods to estimate the trigger efficiency, or to provide correction factors to the MC one; systematic effects associated to the possible methods chosen to compute such trigger efficiency; and a few strategies to keep in mind when proposing a possible new trigger for a future analysis.


Responsible at CMSDAS LPC@FNAL : Aram Avetisyan (Boston U.), Bryan Dahmes (U. of Minnesota), Souvik Das (U. of Florida), Clint Richardson (Boston U.)
Responsible at CMSDAS CERN: Andrea Bocci (CERN), Andrea Perrotta (Bologna)

13. Missing Transverse Momentum (MET) : SWGuideCMSDataAnalysisSchool2014METExerciseFNAL

Description: This is a revived exercise on missing transverse momentum (MET). This short exercise introduces the basics of MET reconstruction and corrections and teaches how to use MET software in CMSSW.


Responsible at CMSDAS LPC@FNAL : Tai Sakuma (TAMU), Markus Stoye (CERN)
Responsible at CMSDAS@CERN : Incorporated into 1 exercise with Jets (see 1)

14. Particle Flow : SWGuideCMSDataAnalysisSchoolPFlowShortExercise

Description: In this exercise, you will learn the basics of particle flow by creating your own "toy" algorithm (which you will compare with the real CMS algorithm). You will not learn how to use particle flow in your analysis; there are many excellent tutorials which already do that. This exercise is designed to teach you what particle flow is. Particle Flow reconstruction aims at reconstructing all stable particles within an event from first principles.


Responsible at CMSDAS LPC@FNAL : Richard Cavanaugh ( FNAL/CHICAGO),Anton Anastassov ( NORTHWESTERN),Sean Kalafut ( MINNESOTA)
Responsible at CMSDAS CERN: Christopher Silkworth ( CHICAGO),Colin Bernet ( CERN)

15. Roostat : SWGuideCMSDataAnalysisSchoolStatistics2014

Description: Introduction to limit setting and statistics tools. Try out using the combine tool and RooStats to produce limits using various statistical techniques and including systematic errors.


Responsible at CMSDAS LPC@FNAL : Jake Anderson ( FERMILAB), John Paul Chou ( RUTGERS)
Responsible at CMSDAS CERN: Josh Bendavid ( CERN),Ted Kolberg ( MARYLAND)

16. Modern Tools for Interactive Analysis : SWGuideCMSDataAnalysisSchoolModernToolsForInteractiveAnalysisExercise

Description: The exercise will give an introduction to pyroot, FWLite, how to perform a fit on data with RooFit embedding the results in the IPython Notebook, and how to use TMVA to study variable correlations and develop multivariate analysis methods. Knowledge used in the examples above is combined to produce a measurement of a trigger efficiency with the tag and probe method. The successful student will be able to develop a documented analysis with pyroot, and produce embedded plots and results in the IPython Notebook.


Responsible at LPC: Marco De Mattia (Purdue University), Zhen Hu (Purdue University)

Long Exercises

Template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT BELOW) - NameOfShortExercises : twiki url


Responsible at CMSDAS LPC@FNAL : FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)
Responsible at CMSDAS CERN: FIRSTNAME LASTNAME ( INSTITUTE),FIRSTNAME LASTNAME ( INSTITUTE)

FNAL: Higgs discovery in 4l and properties SWGuideCMSDataAnalysisSchoolHZZ4lSearchExercise

Description: The 4 leptons final state has been one of the main channel in the discovery of the boson at 125 GeV. The importance of this channel is due to the very small background and very good mass resolution. So, as first step, we will review the analysis flow to understand how we can select a clean signal. We will analyze the kinematics of the signal and the main background (ZZ) and we will extract the expected lepton efficiency and resolution. We will study the features of the reducible background (Z+jets) and we will use control region in data to cross-check the expected number of Z+jets events in the signal region. As second step, we will apply statistical analysis to the results to compute the p-value of the observed excess in data and, if time will permit, we will extract a measurement of the mass of the discovered resonance and the discrimination between spin/parity hypotheses.

Prerequisites: Muons and RooStat short exercise are recommended


Responsibles at CMSDAS LPC@FNAL: Nicola De Filippis ( INFN Bari), Andrew Whitbeck ( FNAL), John Smith ( University California Davis), Predrag Milenovic ( FLORIDA-UNIV)

CERN: Higgs combination and properties SWGuideCMSDataAnalysisSchool2014HiggsCombPropertiesExercise

Description: The purpose of this long exercise is to understand how some of the key properties of the Higgs boson are measured at CMS: how the couplings are determined from the combination of the results of the searches in different final states (decay modes, production topologies), and how exotic JCP hypotheses are tested and excluded from the kinematics of the Higgs boson decay products.

For this exercise, we will use the data from the real higgs searches in CMS, and analyze them with the standard CMS statistical toolkit that has been developed for this purpose.

Prerequisites: basics of statistic and RooFit needed. RooStats short exercise strongly recommended.


Responsible at CMSDAS CERN: Giovanni Petrucciani ( CERN),Pierluigi Bortignon ( ZURICH-ETH), Nicholas Wardle ( LONDON-IC)

CERN: Higgs to 4l SWGuideCMSDataAnalysisSchoolFourLeptonsCERN

Description: The students will be guided to select H->4l candidate events using the 8 TeV dataset. First of all they will optimise the analysis strategy to improve the significance of the signal observation using simulated samples of signal and background events [pre-requisites: muons and ele-photons short exercises] As a second step the data will be "unblinded" and the 4 lepton mass distribution will be statistically interpreted to quantify the significance of the data excess around 126 GeV. The 4-lepton events contributing to the excess will be used to measure some properties of the new boson [pre-requisites: rootstat short exercises]

Prerequisites: Muons, Electrons and RooStat short exercise are recommended


Responsible at CMSDAS CERN: Michalis Bachtis ( CERN), Cristina Botta ( CERN), Predrag Milenovic ( FLORIDA-UNIV)

Higgs to gamma gamma : SWGuideCMSDataAnalysisSchool2014HggExercise

Description: The diphoton channel was instrumental in the discovery of the Higgs-like resonance around 125 GeV. The presence of a very large irreducible background from the non-resonant annihilation of quarks into two photons makes this a challenging analysis. The primary handle used to differentiate signal from background is a good diphoton invariant mass resolution, and for this reason the use of multivariate techniques to optimize this resolution are critical to the analysis. In this exercise we will train a boosted decision tree to improve the photon energy resolution (and hence the diphoton mass resolution), practice fitting using RooStats, and use the Higgs combination tool to perform a shape analysis on the result.


Responsible at CMSDAS LPC@FNAL : Rishi Patel (Rutgers), Marc Weinberg (Florida State)
Responsible at CMSDAS CERN: Josh Bendavid ( CERN),Louis Sgandurra ( IPN LYON), Marco Peruzzi ( ZURICH-ETH)

Higgs to bbar https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolHiggsbbbar2014Exercise

Description: The observation of a Higgs-boson like in 2012 with a mass of 126 GeV has been an extraordinary event for CMS and for particle physics. The next goal of the experiment is now to characterize the properties of this new boson, and confirm/exclude that is behaving as the Standard Model (SM) Higgs. One fundamental property to study is the new boson coupling to fermions. We will focus on a H->bb search which has the largest decay branching ratio (~60% at 125 GeV). In this exercise we will start with an overview of the differences between signal and background processes and corresponding discriminating variables in the channel. We will try to define our selection cuts in order to maximize the probability to observe a signal in our data. We will perform a fit to the dijet invariant mass shape to set limits on the Higgs cross section (or observe an excess compatible with the SM Higgs) and we will look into several systematic uncertainty contributions (like b-tag efficiency, jet resolution,etc.). We will perform a dedicated analysis for the di-boson process W(Z)+ Z->bb, that provides a crucial calibration for the Higgs to bbar search. The methods used in the exercises will be very similar to the ones used in the official analysis.


Responsible at CMSDAS LPC@FNAL : Souvik Das, Jia Fu Low (UF), Leonard Apanasevich (UIC), Caterina Vernieri (Pisa)
Responsible at CMSDAS CERN: Not done at CERN

SUSY Hadronic

Description: This analysis is a generic search for new physics. It is motivated by models of R-parity conserving Supersymmetry (SUSY) that generally assume the existence of additional elementary particles. The SUSY signature we are looking for in the detector consists of several jets with high pT, large missing transverse momentum due to the LSPs, and no leptons. The data-based determination of the SM backgrounds is one of the key features of this analysis. This exercise will follow the recently published analysis of the 19.5/fb of data collected in 2012 at a center-of-mass energy of 8 TeV. The focus of the exercise will be on understanding (some of) the data-based background determination methods which are the heart of this analysis.


Responsible at CMSDAS LPC@FNAL : Hongxuan Liu (Baylor), Keith Ulmer (Colorado), Markus Stoye (CERN), Benjamin Stieger (ETH Zurich), Azeddine Kasmi (Baylor), Daryl Hare (Fermilab), Seema Sharma (Fermilab)
Responsible at CMSDAS CERN : Filip Moortgat (CERN), Matthias Schröder (DESY), Jared Sturdy (University of California, Riverside)

Exotica exercise Z'--> dilepton (mumu): https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2014DimuonExercise

Description: This extended exercise is intended to familiarize you with two dimuon analysis efforts in CMS. The Drell-Yan to dimuon spectrum is a fruitful signature for discovering new physics, and we seek to measure the Drell-Yan spectrum to high precision in CMS. Many new models of physics predict a new narrow resonance, here referred to generically as Z', that decays into a pair of oppositely-charged muons. We look for a bump from Z' in the dimuon invariant mass spectrum using a shape-based search, looking for the difference in between the exponentially falling background and the resonance peak that would show up should a Z' exist. In this exercise, you'll work on the meat of the analysis to produce the dimuon mass spectra for data and MC. In doing this exercise you can learn (among other things): how to apply some loose and tight selection criteria to select muons and make combined dimuon object collections; how to evaluate acceptance, trigger efficiency, mass resolution; how to estimate some of the backgrounds using the data itself.

Prerequisites: Muons short exercise are recommended


Responsibles at CMSDAS LPC@FNAL: Hwidong Yoo (Purdue University), Bryan Dahmes (University of Minnesota), Greg Landsberg (Brown University)

Jets (Exotica with Jets): SWGuideCMSDataAnalysisSchoolJetsLongExerciseDijetMassSpectrumInAssociationWithW

Description: The long exercise will teach you how to search for new physics with jets by looking at the dijet mass spectrum in association with the W boson. You will study the physics of the dijet spectrum, analyze relevant processes, learn to recognize their signatures and develop analysis techniques while examining CMS Data.

It is strongly recommended that students take the Jet Short Exercise prior to taking the Jet Long Exercise. Other related short exercises are pile-up, particle flow, and generators. Limits will not be extracted in the jet long exercise, so the Roostats short exercise is a valuable supplement to the jet long exercise for those students planning to set limits in the future.

Prerequisites: There are no requirements, but students are strongly encouraged to take the jets short exercise ( SWGuideCMSDataAnalysisSchoolJetAnalysis) first.


Responsible at CMSDAS LPC@FNAL : Ajay Kumar ( University of Delhi ), Alexx Perloff (Texas A&M), John Stupak (Purdue-Calumet), Zhenbin Wu (Baylor )
Responsible at CMSDAS CERN:

Exotica boosted dijets (Search for RS gravition decaying to WW in the dijet mass spectrum) : SWGuideCMSDataAnalysisSchool2014EXODijetsLongExercise


Responsible at CMSDAS LPC@FNAL : Nhan Tran (FNAL), Tulika Bose (BU), James Dolen (Buffalo), Joshua Swanson (Wisconsin)
Responsible at CMSDAS CERN: Konstantinos Kousouris ( CERN),Maxime Gouzevitch( IPN-LYON), Aram Avetisyan( BOSTON-UNIV)

Top quark cross-section and top-like BSM searches https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolTopExercise

Measurement of the top quark pair production cross section in events containing lepton+jets using b-quark identification and top-like BSM searches.

Description: The basic exercise measures the top quark pair production cross section using events containing one isolated muon, 3 or more jets, missing transverse energy, indicative of the presence of a neutrino, and using b-quark identification. This sample can also be used to search for new physics if there is interest in doing so.

Students will learn about the main background sources to the top quark signal in this channel, and use a data-driven method to estimate the QCD multijet background. The top quark pair production cross section will be measured from a fraction of the 2012 data using two methods: template fit and counting. In addition, the students will learn how to search for heavy top quark resonances.

Suggested short exercises: b-tagging, muons, particle flow, RooStats (strongly suggested).


Responsible at CMSDAS@LPC: Freya Blekman ( Brussels), Sadia Khalil ( KSU), Francisco Yumiceva( Florida Tech), Susan Dittmer ( Cornell)
Responsible at CMSDAS CERN: Freya Blekman ( Brussels),Jim Dolen ( SUNY-BUFFALO), James Keaveney (Brussels)

FNAL: Exotica displaced v. https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2012ExoticaDisplacedVertices

Description: In this exercise, you will perform a search for a heavy resonance (M) decaying to two long-lived massive neutral particles (m), which then decay to leptons (electrons and muons). The full chain is M -> m m -> 4 leptons, with both pairs of leptons being displaced from the primary vertex. This highly exotic signature would be clear evidence of physics beyond the Standard Model, and is motivated by hidden valley scenarios.

You will perform the complete analysis with full statistics. The exercise will cover MC simulation, kinematics, definition of acceptance, efficiencies, and will use the RooStats tools to set upper limits (unless we see something new). We will discuss and explore sources of systematic uncertainty and how to estimate them using data-driven methods wherever possible.

The exercise will be based on 2011 and 2012 data. The increased energy and statistics will lead to an increased sensitivity and provide a first look beyond the existing limits which were derived on 2011 data.

Preferred short exercises: tracking, muons, Btag & vertexing, RooStats.


Responsible at CMSDAS@LPC: Marco De Mattia ( Purdue), Zhen Hu ( Purdue)

Heavy Ions : https://twiki.cern.ch/twiki/bin/view/CMS/ChargedSpectraDAS2014CERN


Responsible at CMSDAS LPC@FNAL : Not done at FNAL
Responsible at CMSDAS CERN: Gabor Veres ( CERN),Lamia Benhabib ( CERN), Krisztian Krajczar ( CERN)

Standard Model : https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolZandWinclusiveExercise

Inclusive Z and W cross section measurement

Description: The purpose of this long exercise is to build a foundation for performing analyses at a hadron collider detector. We will do this by starting from the inclusive W and Z production. The final goal is to teach students how to measure the integrated cross section. The basic physics of the vector boson production, relevant processes and their signatures will be explained. The analysis techniques used to examine the data will be developed and explored. We will use the technical experience gained in the pre-exercises and short-exercises as a basis for the long exercise. The purpose of the long exercise is to emphasize analysis techniques, rather than strictly software machinery. Of course, we will continue to develop and hone the machinery as we go along, but we will treat CMS software and ROOT as a means rather than an end. The course design is intended for starting graduate students with little or no prior experience in high energy physics analysis techniques.

(Strongly) Suggested short exercises: muons, electron-photons, generators


Responsible at CMSDAS LPC@FNAL : Not done at FNAL
Responsible at CMSDAS CERN: Pedro Silva ( CERN),Phil Harris ( CERN), Martijn Mulders ( CERN), Luca Perrozzi ( CERN), Benjamin Stieger( CERN)

CMSDAS 2013 (archive) (LPC, DESY, SINP)

Pre-Workshop Exercises

Short Exercises

Template (DO NOT EDIT THIS BUT CUT PASTE AND USE IT BELOW) - NameOfShortExercises : twiki url

Description: CUT PASTE DESCRIPTION HERE


Responsible at CMSDAS LPC@FNAL : FIRSTNAME LASTNAME ( INSTITUTE),
Responsible at CMSDAS Terascale@DESY: FIRSTNAME LASTNAME ( INSTITUTE),
Responsible at CMSDAS SINP@Kolkata: FIRSTNAME LASTNAME ( INSTITUTE),

1. Jets : SWGuideCMSDataAnalysisSchoolJetAnalysis

Description: The short exercise provides hands-on experience accessing jets, plotting basic jet quantities, and teaches the basics of jet energy corrections and their uncertainty. You will learn the various available pile-up corrections for jets. You will become familiar with basic jet types and algorithms, including jet sub-structure algorithms.


Responsible at CMSDAS LPC@FNAL : Robert Harris ( FNAL), Kalanand Mishra ( FNAL), Ilya Osipenkov ( TAMU)
Responsible at CMSDAS @ Hamburg: Roman Kogler (Hamburg ),
Responsible at CMSDAS @ Kolkata: Paolo Gunnellini ( DESY), Gobinda Majumdar ( TIFR), John Stupak ( Purdue Calumet)

2. Electrons : SWGuideCMSDataAnalysisSchool2013ElectronExercise

Description: The short exercise provides hands-on experience with electron identification at CMS. A participant will learn how to analyze simple TTree files and apply cut-based identification criteria to select Z->ee events in data and MC simulation. A participant will also be given examples on how to access all the variables needed to identify electrons with 7 and 8 TeV CMS runs.


Responsible at CMSDAS LPC@FNAL : Yurii Maravin ( KSU), Irakli Svintradze ( KSU)
Responsible at CMSDAS @ Hamburg: Hannes Schettler ( Hamburg U.)

3. Photons : SWGuideCMSDataAnalysisSchool2013PhotonExercise

Description: The short exercise provides hands-on experience with photon identification at CMS. A participant will learn how to analyze simple TTree files and apply cut-based identification criteria to select events in data and MC simulation. A participant will also be given examples on how to access all the variables needed to identify photons.


Responsible at CMSDAS LPC@FNAL : Anthony Barker ( Rutgers), Yurii Maravin ( KSU)
Responsible at CMSDAS @ Hamburg: Valentina Sola ( Hamburg U.)
Responsible at CMSDAS @ Kolkata: Andrew Askew ( FSU), Yurii Maravin ( KSU)

4. Muons : SWGuideCMSDataAnalysisSchool2013MuonExercise

Description: This exercise is intended to familiarize you with using muons in a physics analysis. In particular, you will learn: a brief overview of muon reconstruction and identification algorithms, what muon related information is available in AOD and PAT and how to access muon specific information that could be relevant for your analysis. We will show you what typical performance you can expect for muons in CMS, regarding efficiency, acceptance, purity and momentum resolution. The exercise will also cover the basics of measuring the muon identification and trigger efficiency in data.


Responsible at CMSDAS LPC@FNAL : Martijn Mulders ( CERN) and Nicola De Filippis ( INFN Bari)
Responsible at CMSDAS @ Hamburg: Ivan Asin, Carmen Diez Pardos and Francesco Costanza ( DESY)
Responsible at CMSDAS @ Kolkata: Frank Golf ( Univ. of California Santa Barbara, US), Luca Perrozzi ( CERN)

5. Particle Flow https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolPFlowShortExercise

Description: In this exercise, you will learn the basics of particle flow by creating your own "toy" algorithm (which you will compare with the real CMS algorithm). Particle Flow reconstruction aims at reconstructing all stable particles within an event from first principles. These particles are then used as input to reconstruct more higher-level, derived objects such as taus, jets, missing transverse momentum...just as could be done with true particles from a simulation. The basic building elements for the particle flow algorithm are tracks, electromagnetic clusters and hadronic calorimeter clusters.

The goal of this exercise is to introduce the very basic concepts of particle flow event reconstruction. After you finish this exercise, you should have a basic understanding of what particle flow is and how (and why) it works, especially in the case of CMS. Historically, the CMS particle flow algorithm was developed by first scanning events and visualizing how different types of individual particles interact in across different CMS sub-detectors. The CMS Particle Flow Algorithm is documented at SWGuideParticleFlow.


Responsible at CMSDAS @ Hamburg: Adrian Perieanu ( RWTH Aachen), Alexei Raspereza (DESY)
Responsible at CMSDAS @ Kolkata: Christopher Silkworth ( University of Illinois at Chicago, US)

6. Pile-up https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2013PileupReweighting

Description: Learn the procedure for reweighting MC distributions to match the pileup in the data. Students will learn how to access the various pieces of information related to pileup in the MC events, and how to use the reweighting tools.

Responsible at CMSDAS@LPC: Mike Hildreth ( Notre Dame)
Responsible at CMSDAS @ Hamburg: Dirk Kruecker ( DESY)
Responsible at CMSDAS @ Kolkata: Nitish Dhingra ( Panjab University, IN), Sudhir Malik ( University of Nebraska, US)

7. Visualization https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2012EventScanningExercise

Description: The goal of this short exercise is to enable the student to “scan” interesting CMS events. The tool to be used is Fireworks. This tool has been used in early cosmic ray events (CRAFT) and then in early running at 0.45 and 2.36 TeV. The representative data used here is from the period April – October 2010 when the LHC luminosity was doubling every~ 2 weeks. During that period an “Exotica Scan” was set up, which selected the ~ 30 most bizarre events each day and asked a small group of “scanners” to try to understand them. The goal was to identify problems and issues with the CMS detector. The exerises of is of general use when looking at events even at 8 TeV or at higher energy in future era.


Responsible at CMSDAS@LPC and @ Hamburg: Sudhir Malik ( Nebraska/FNAL), Alexey Ferapontov ( Brown University)
Responsible at CMSDAS @ Kolkata: Devdatta Majumder ( National Taiwan University, TW)

8. Tracking and Primary Vertices SWGuideCMSDataAnalysisSchool2013TrackingExercise

Description:

We will present an introduction to using tracks for analyses in the era of large pile-up (many primary vertices). Our exercises will all use real data and will familiarize you with the following techniques:

  • Extracting basic track parameters and reconstructing invariant masses from tracks in CMSSW. Tracks are the closest detector entity to the four-vectors of particles: the momentum of a track is nearly the momentum of the charged particle itself.
  • Cleaning sets of tracks for analysis. We will use POG-endorsed filters to eliminate bad tracks and discuss sources of tracking uncertainties.
  • Extracting basic parameters of primary vertices. In this high-luminosity era, it is not uncommon for a single event to contain as many as ten to twenty independent collisions. For most analyses, only one is relevant, and it can usually be identified by its tracks.


Responsible at CMSDAS @ Hamburg : Nazar Bartosik ( DESY), Joerg Behr ( DESY), Gregor Hellwig ( DESY)
Responsible at CMSDAS LPC@FNAL : Kevin Burkett ( Fermilab), Nhan Viet Tran ()
Responsible at CMSDAS @ Kolkata: Giovanni Petrucciani ( CERN), Sanjay Padhi ( UCSD)

9. B-tagging and vertexing https://twiki.cern.ch/twiki/bin/viewauth/CMS/SWGuideCMSDataAnalysisSchoolbTagExercise

Description:

We will learn about the properties and differences between primary vertices and the luminous region (also known as beam spot). Students will learn how to access the vertex collections and make plots of the main variables. Furthermore, they will learn how to identify b-flavored jets and plot the b-tag efficiency and mistagging rate as a function of jet kinematics for a given algorithm. Use the b-tagging scale factors derived from data to correct the Monte Carlo efficiencies.


Responsible at CMSDAS @ LPC@FNAL: Francisco Yumiceva (FNAL), Pratima Jindal(Princeton)
Responsible at CMSDAS @ Hamburg: Roberval Walsh (DESY)
Responsible at CMSDAS @ Kolkata: Francisco Yumiceva ( Florida Institute of Technology, US), Mariarosaria D'Alfonso ( CERN)

10. Generators https://twiki.cern.ch/twiki/bin/viewauth/CMS/SWGuideCMSDataAnalysisSchool2013GeneratorExerciseatSINP

Description:

This exercise is based on a longer tutorial developed for the Hadron Collider Physics Summer School (HCPSS) which would work outside of CMSSW. The current tutorial is shorter, but works within CMSSW. For those who are interested, the notes and software guide from the HCPSS are duplicated here. This tutorial will guide you on an investigation of the similarities and differences that can be encountered when comparing predictions made with different Monte Carlo (MC) tools. In all cases, we focus on W+ and W- production and decay to leptons at the LHC (a 7 TeV, proton-proton collider).


Responsible at CMSDAS @ DESY: Altan Cakir (DESY)
Responsible at CMSDAS @ Kolkata: Altan Cakir (DESY), Tyler Dorland(DESY)

11. RooStats : SWGuideCMSDataAnalysisSchoolStatisticsFall2012

Description: The goal of the exercise is to get acquainted with the statistics analysis practices used in CMS. This includes the conceptual understanding of the models and methods, and familiarity with popular computing tools. The exercise emphasizes the use of the CombinedLimit tool (also known as the Higgs tool). CombinedLimit and other tools use the RooFit /RooStats framework for calculations. The exercise also demonstrates how to use the RooStats framework directly.


Responsible at CMSDAS LPC@FNAL : Gena Kukarsev (Brown University/FNAL), Guillermo Breto Rangel (UC Davis)
Responsible at CMSDAS Terascale@DESY: Jochen Ott (KIT),
Responsible at CMSDAS @ Kolkata: Jake Anderson ( FNAL)

12. HLT : SWGuideCMSDataAnalysisSchool2013HLTExercise

Description: The short exercise is intended to introduce the student with the methods used to retrieve trigger related informations, and to access them from within an analyzer, in view of using those informaions to estimate the trigger efficiency relevant for any given analysis. In particular, one will learn: how to find which trigger menu was used for a given dataset; how to browse through the HLT paths in that menu and inspect the content of the modules, filters and producers, therein; the roles of the different filters in the path; understand which objects are saved by a HLT filter, and how to access them. In the final discussion we will use what learned in the exercise and try to elaborate about data driven methods that can be applied to estimate the trigger efficiency, or to provide correction factors to the MC one, systematic effects associated to the possible method chosen to compute the trigger efficiency, and possible strategies to keep in mind when proposing a new trigger for a future analysis


Responsible at CMSDAS @ Kolkata: Andrea Perrotta ( INFN Bologna)

13. Modern Tools for Interactive Analysis : SWGuideCMSDataAnalysisSchoolModernToolsForInteractiveAnalysisExercise

Description: The exercise will give an introduction to pyroot, FWLite, how to perform a fit on data with RooFit embedding the results in the IPython Notebook, and how to use TMVA to study variable correlations and develop multivariate analysis methods. Knowledge used in the examples above is combined to produce a measurement of a trigger efficiency with the tag and probe method. The successful student will be able to develop a documented analysis with pyroot, and produce embedded plots and results in the IPython Notebook.


Responsible at CMSDAS SINP@Kolkata: Marco De Mattia ( Purdue University)

Long Exercises ( Under update for the CMSDAS 2013 )

Description: CUT PASTE DESCRIPTION HERE
-->

1. Jets (new physics): SWGuideCMSDataAnalysisSchoolJetsLongExerciseDijetMassSpectrumInAssociationWithW

Description: The long exercise will teach you how to search for new physics with jets by looking at the dijet mass spectrum in association with the W boson. You will study the physics of the dijet spectrum, analyze relevant processes, learn to recognize their signatures and develop analysis techniques while examining CMS Data.

It is strongly recommended that students take the Jet Short Exercise prior to taking the Jet Long Exercise. Other related short exercises are pile-up, particle flow, and generators. Limits will not be extracted in the jet long exercise, so the Roostats short exercise is a valuable supplement to the jet long exercise for those students planning to set limits in the future.

Prerequisites: There are no requirements, but students are strongly encouraged to take the jets short exercise ( SWGuideCMSDataAnalysisSchoolJetAnalysis) first.


Responsible at CMSDAS LPC@FNAL : Robert Harris ( FNAL), Kalanand Mishra ( FNAL), Ilya Osipenkov ( TAMU)
Responsible at CMSDAS @ Hamburg: Sebastian Naumann (DESY)
Responsible at CMSDAS @ Kolkata: Paolo Gunnellini ( DESY), John Stupak ( Purdue Calumet), Luca Perrozzi ( CERN), Manoranjan Guchait ( TIFR)

2. Photons (A peek inside Higgs->gamma gamma analysis) : Exercise Twiki (under construction)

Description:

Pre-requiste short-exercises: Photons and RooStat are recommended

The di-photon channel was one of the instrumental channels in Higgs discovery. Attempting the entire analysis chain in one exercise is quite impossible. But we can get a good understanding of the heart of the analysis: a tandem of two regressions, one to improve the photon energy resolution, and the other to measure its performance and classify the events according to how well-measured the Higgs candidates are. You will learn something about significance measurements, too.


Responsible at CMSDAS LPC@FNAL : David Mason (FNAL), Andrew Askew (FSU), Yuri Gershtein (Rutgers)
Not performed at CMSDAS @ Hamburg!

Responsible at CMSDAS @ Kolkata: Andrew Askew ( FSU), Yurii Maravin ( KSU), Satyaki Bhattacharya ( SINP)

3. Supersymmetry : Exercise Twiki

Description:

Models of supersymmetry (SUSY) that conserve R-parity and include a neutral, weakly interacting lightest SUSY particle (LSP) garner much interest because they simultaneously solve the hierarchy problem, allow unification of the fundamental interactions, and provide a candidate for dark matter. Many searches for SUSY focus on the presence of large missing transverse energy (MET) carried away by the LSP. This approach neglects well motivated SUSY models that predict low MET without special tuning of masses -- models characterized by R-parity violation, gauge mediated SUSY breaking, compressed spectra, and hidden valleys. As the parameter space available for high-MET SUSY is reduced by recent results from the LHC, low-MET alternatives become more important to study.

We will perform a general search for new particles decaying in a cascade to two photons, several jets, and low MET. We will interpret the search in terms of the new "stealth" SUSY model, which predicts low-MET signatures while conserving R-parity and without special tuning of masses.

You will learn:

  • Basic theory of SUSY and its low-MET variants including stealth SUSY,
  • A simple and robust method for estimating background from the data,
  • Fitting data with RooFit in the PyROOT framework,
  • Optimization of selection criteria,
  • Estimation of systematic uncertainty, and
  • Limit setting techniques.

Prerequisites: There are no firm prerequisites, but students would benefit from attending the jet, photon, or RooStats /RooFit short exercises.


Responsible at CMSDAS LPC@FNAL : Jim Hirschauer (FNAL), Daniel Elvira (FNAL), Ken Hatakeyama (Baylor), Ben Hooberman (FNAL), and Manfred Paulini (Carnegie Mellon).
Not performed at CMSDAS @ Hamburg.

4. Higgs in 4 leptons SWGuideCMSDataAnalysisSchoolHZZ4lSearchExercise

Description: The 4 leptons final state has been one of the main channel in the discovery of the boson at 125 GeV. The importance of this channel is due to the very small background and very good mass resolution. So, as first step, we will review the analysis flow to understand how we can select a clean signal. We will analyze the kinematics of the signal and the main background (ZZ) and we will extract the expected lepton efficiency and resolution. We will study the features of the reducible background (Z+jets) and we will use control region in data to cross-check the expected number of Z+jets events in the signal region. As second step, we will apply statistical analysis to the results to compute the p-value of the observed excess in data and, if time will permit, we will extract a measurement of the mass of the discovered resonance.

Prerequisites: Muons and RooStat short exercise are recommended


Responsibles at CMSDAS LPC@FNAL: Nicola De Filippis ( INFN Bari), Ian Anderson and Andrei Gritsan ( Johns Hopkins U), Martijn Mulders ( CERN), John Smith ( University California Davis)
Responsibles at CMSDAS @ Hamburg: Nicola De Filippis and Piet Verwilligen ( INFN Bari), Giacomo Ortona( Univ. of Torino), Niklas Pietsch and Martin Goerner( DESY)
Responsible at CMSDAS @ Kolkata: Nicola De Filippis and Simranjiit Singh Chhibra ( INFN Bari), Chia Ming Kuo ( NCU)

5. Higgs to bbar https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolLowMassHiggsSearch2013Exercise

Description: The observation of a Higgs-boson like in 2012 with a mass of 126 GeV has been an extraordinary event for CMS and for particle physics. The next goal of the experiment is now to characterize the properties of this new boson, and confirm/exclude that is behaving as the Standard Model (SM) Higgs. One fundamental property to study is the new boson coupling to fermions. We will focus on a H->bb search which has the largest decay branching ratio (~60% at 125 GeV). In this exercise we will start with an overview of the differences between signal and background processes and corresponding discriminating variables in the channel. We will try to define our selection cuts in order to maximize the probability to observe a signal in our data. We will perform a fit to the dijet invariant mass shape to set limits on the Higgs cross section (or observe an excess compatible with the SM Higgs) and we will look into several systematic uncertainty contributions (like b-tag efficiency, jet resolution,etc.). We will perform a dedicated analysis for the di-boson process W(Z)+ Z->bb, that provides a crucial calibration for the Higgs to bbar search. The methods used in the exercises will be very similar to the ones used in the official analysis.


Responsibles at CMSDAS LPC@FNAL: David Lopes Pegna ( Princeton/Purdue), Michele de Gruttola ( Univ of Florida)
Responsible at CMSDAS @ Hamburg: Dean Horton ( DESY), Holger Enderle ( University of Hamburg), Elias Ron ( DESY), Ozgur Sahin ( DESY)
Not performed at CMSDAS @ Kolkata.

Pre-requisite short-exercises: Particle Flow, Statistics, B-jets, Jets

6. Physics with taus SWGuideCMSDataAnalysisSchool2013PhysicsWithTausExercise

Description: "Taus are priviliged probes of new physics. In the context of the CMS Higgs boson(s) searches, final states with taus are well established and promising channels. With the whole of 2011 data set, the Higgs to tau-tau channel alone can set a limit on the Higgs boson cross-section from 3 to 5 times the SM prediction in the range 115-145 GeV. This has been achieved thanks to the combination of multiple tau final states and by exploiting the different production modes (gg-fusion, vector boson fusion) as to maximize the statistical power of the full data set. In models with extended Higgs sectors, like the MSSM, the coupling of the Higgs boson(s) to taus can be significantly enhanced depending on the model parameters: this makes the tau lepton a probe for supersymmetry. In this exercise, a simple cut-based analysis is proposed to isolate a clean sample of events where a Z boson decays to a pair of taus, which in turn decay to a muon (plus neutrinos) and pions (plus neutrino). The students will be asked to produce the visible mass spectrum of the tau-pair, which noticeably exhibits a clean Z peak. The basic steps of a search analysis will be presented: the SM backgrounds will be calibrated in control regions pretty much as done in the official CMS analysis; a set of data/MC correction factors will be proposed to correct the MC efficiency. Depending on the time, the results of the analysis will be used as a simple counting-experiment to set an upper limit on the SM Higgs boson cross-section."
Preferred short exercises: Particle-Flow, Muon.


Responsible at CMSDAS LPC@FNAL: Jeff Kolb ( Notre Dame), Keti Kaadze ( CERN), Mike Hildreth( Notre Dame)
Responsible at CMSDAS @ Hamburg: Armin Burgmeier ( DESY), Adrian Perieanu ( RWTH)
Not performed at CMSDAS @ Kolkata.

7. Supersymmetry (Jets + MHT) :

@ SINP, Kolkata SWGuideCMSDataAnalysisSchool2013SUSYJetsPlusMHTSINP

This analysis is a generic search for new physics. It is motivated by models of R-parity conserving Supersymmetry ( SUSY) that generally assume the existence of additional elementary particles. In many SUSY models, gluino-gluino, gluino-squark, and squark-squark pair production can occur with particularly large cross sections compared to other SUSY-production channels. The squarks and gluinos will predominantly decay into coloured SM particles and pairs of stable LSPs ( Lightest Supersymmetric Particles). In realistic models, the LSPs are electrically neutral and only weakly interacting such as neutrinos. Thus, the SUSY signature we are looking for in the detector consists of several jets with high pT, large missing transverse momentum due to the LSPs, and no leptons.

This exercise will follow the recently published analysis of the 19.5/fb of data collected in 2012 at a center-of-mass energy of 8 TeV. The focus of the exercise will be on understanding (some of) the data-based background determination methods which are the heart of this analysis.


Responsible at CMSDAS @ Kolkata: Matthias Schroeder ( DESY ), Altan Cakir ( DESY )

@ DESY, Hamburg SWGuideCMSDataAnalysisSchool2013SUSYJetsPlusMHT

This analysis is a generic search for new physics. It is motivated by models of R-parity conserving Supersymmetry ( SUSY), which predict the production of gluino-gluino, gluino-squark, and squark-squark pairs in the proton-proton collisions at the LHC via the strong interaction. In many models, these reactions have a particularly large cross section compared to other SUSY-production channels. The squarks and gluinos will predominantly decay into coloured SM particles and pairs of stable LSPs ( Lightest Supersymmetric Particles). In realistic models, the LSPs are electrically neutral and only weakly interacting such as neutrinos. Thus, the SUSY signature we are looking for in the detector consists of several jets with high pT, large missing transverse momentum due to the LSPs, and no leptons.

The expected rate of SM processes with the same signature ( SM background) is determined almost entirely from data. This is beneficial because the reliance on the simulation - and with it the uncertainity on the background rate - is minimised. The data-based determination of the SM backgrounds is one of the key features of this analysis.

For this exercise, we will follow the currently ongoing analysis of the 8 TeV data taken in 2012. We will use the first 5/fb of 2012 data. The focus of the exercise will be on understanding the data-based background determination methods.


Responsible at CMSDAS Terascale@DESY: Altan Cakir ( DESY), Francesco Costanza ( DESY), Hannes Schettler ( DESY), Matthias Schröder ( University of Hamburg)

8. Exotica displaced v. https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2012ExoticaDisplacedVertices

Description: In this exercise, you will perform a search for a heavy resonance (M) decaying to two long-lived massive neutral particles (m), which then decay to leptons (electrons and muons). The full chain is M -> m m -> 4 leptons, with both pairs of leptons being displaced from the primary vertex. This highly exotic signature would be clear evidence of physics beyond the Standard Model, and is motivated by hidden valley scenarios.

You will perform the complete analysis with full statistics. The exercise will cover MC simulation, kinematics, definition of acceptance, efficiencies, and will use the RooStats tools to set upper limits (unless we see something new). We will discuss and explore sources of systematic uncertainty and how to estimate them using data-driven methods wherever possible.

The exercise will be based on 2011 and 2012 data. The increased energy and statistics will lead to an increased sensitivity and provide a first look beyond the existing limits which were derived on 2011 data.

Preferred short exercises: tracking, muons, Btag & vertexing, RooStats.


Responsible at CMSDAS@LPC: Jim Pivarski ( Fermilab),Ian Shipsey ( Purdue), Marco De Mattia ( Purdue)
Responsible at CMSDAS @ Hamburg: Nuno Leonardo ( Purdue), Ian Shipsey ( Purdue), Marco De Mattia ( Purdue), Alexander Schmidt ( DESY)
Responsible at CMSDAS @ Kolkata: Marco De Mattia ( Purdue University), Devdatta Majumder ( NTU)

9. Top and top-like searches https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolTopExercise

Measurement of the top quark pair production cross section in events containing lepton+jets using b-quark identification

Description: The basic exercise measures the top quark pair production cross section using events containing one isolated muon, 3 or more jets, missing transverse energy, indicative of the presence of a neutrino, and using b-quark identification. This sample can also be used to search for new physics if there is interest in doing so.

Students will learn about the main background sources to the top quark signal in this channel, and use a data-driven method to estimate the QCD multijet background. The top quark pair production cross section will be measured from the 2011 data using two methods: template fit and counting.

Suggested short exercises: b-tagging, muons, PFlow. Essential: RooStats (fitting part)


Responsible at CMSDAS@LPC: Freya Blekman ( Brussels), James Dolen ( Buffalo), Sadia Khalil ( KSU), Salvatore Rappoccio ( Buffalo), Francisco Yumiceva( Fermilab)
Responsible at CMSDAS @ Hamburg: : Freya Blekman ( Brussels), Tyler Dorland ( DESY), Sadia Khalil ( KSU)
Responsible at CMSDAS @ Kolkata: Francisco Yumiceva ( Florida Institute of Technology, US), Dr. James Michael Keaveney ( VUB - Interuniversitary Institute for high energies (IIHE), BE), Gobinda Majumder ( TIFR)

10. Photons - new physics : https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2013PhotonLongExercise


Description: New physics with vector-like confinement possesses a rich phenomenology, including potentially massive stable charged particles, and final states with multiple high energy photons. In our case, two new particles are created in collision, one decays to two photons, the other decays into a photon and a W boson. Which means in our final state we are guaranteed three high energy photons, two of which are resonant. The challenges will be: Efficiently selecting these events within the CMS detector acceptance, estimating the efficiency times acceptance for this selection, estimating the background (non-resonant) from jets and other processes, setting an upper limit (or potentially n-sigma evidence) based on the background and observed events.


Responsible at CMSDAS Terascale @ Hamburg: Christian Autermann (RWTH Aachen), Valentina Sola (Hamburg)

11. Higgs Properties : https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2013HiggsPropertiesExercise

Description: The purpose of this long exercise is to understand how some of the key properties of the Higgs boson are measured at CMS: how the couplings are determined from the combination of the results of the searches in different final states (decay modes, production topologies), and how exotic JCP hypotheses are tested and excluded from the kinematics of the Higgs boson decay products.

For this exercise, we will use the data from the real higgs searches in CMS, and analyze them with the standard CMS statistical toolkit that has been developed for this purpose.

Prerequisites: basics of statistic and RooFit needed. RooStats short exercise strongly recommended.


Responsible at CMSDAS @ Kolkata : Giovanni Petrucciani ( CERN), Sanjay Padhi ( UCSD), Ashok Kumar ( Univ. of Delhi), Nayeemuddin ( Univ. of Delhi)

12. Heavy Ion - Upsilon https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2013UpsilonExercise

Description: In this exercise we will measure the ratios of yields of Y(2S) and Y(1S) in pp and PbPb collisions.The motivation for such a measurement is to know the modification of the ratio of the yields in nuclear medium. Since the binding energies of the Upsilon states are different the medium is expected to act differently on different states. The exercise is similar to the work presented in the paper arXiv with the corresponding analysis link HIN-11-011. We will use latest pp data set measured in 2013 with integrated luminosity 5.4 pb-1. The PbPb data set measured in 2011 with luminosity 150 μb-1 will be used in the analysis. First we will start with pp data to obtain the yields of Y(2S) and Y(1S) and their ratio. We will then use MC to get the ratio of acceptances and efficiencies of Y(2S) and Y(1S). The yields and ratios will be obtained in PbPb system for few centralities.The double ratio i.e. the ratio of ratios of two states in PbPb and pp systems will be obtained.


Responsible at CMSDAS @ Kolkata: Prashant Shukla ( BARC), Dipanwita Dutta ( BARC)

13. Photons (H->gamma gamma) https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2013HggExercise

Description: The di-photon channel was one of the instrumental channels in Higgs discovery. Attempting the entire analysis chain from scratch in a couple of days is quite impossible. But we can get a good understanding of the heart of the analysis: a tandem of two regressions, one to improve the photon energy resolution, and the other to measure its performance and classify the events according to how well-measured the Higgs candidates are. You will learn something about significance measurements, too. This exercise will not cover identification and selection of photons, which have been covered (at some level) in the short exercise. We will also ignore, for the sake of time and clarity, a lot of important parts of the analysis. For example, we will not cover the primary vertex selection. The basic principles for it are similar to the ones for photon energy: a tandem of regressions, first to improve the measurement (in primary vertex case - pick the correct higgs production vertex) and the second to statistically evaluate its performance (calculate the probability that the vertex is picked correctly). Rather then cramming all necessary parts of the analysis into this one exercise, we'd rather have the students master the underlying principle and have time to play with it.


Responsible at CMSDAS @ Kolkata : Satyaki Bhattacharya ( SINP), Dr. Andrew Warren Askew ( FSU), Yurii Maravin ( KSU)

Exercises Readiness for CMSDAS at LPC@FNAL

Pre-exercises readiness

Exercise Set Responsible(s) Work in progress Ready COMMENTS
Set 1 Sudhir Malik ( UNL/FNAL)   %OK% Announced on 2 Aug 2012
Set 2 Sudhir Malik ( UNL/FNAL)   %OK% "
Set 3 Sudhir Malik ( UNL/FNAL)   %OK% "
Set 4 Sudhir Malik ( UNL/FNAL)   %OK% "

SHORT exercises readiness

SHORT Exercise name Responsible(s) Work in progress Ready (tested at FNAL) COMMENTS
Jets Robert Harris ( FNAL), Kalanand Mishra ( FNAL), Ilya Osipenkov ( TAMU)   %OK% Announced on 18 December 2012
Generators        
Tracking and Primary Vertices        
Electrons Yurii Maravin ( KSU), Irakli Svintradze ( KSU)   %OK%  
Muons Martijn Mulders ( CERN), Nicola De Filippis ( INFN Bari)   %OK%  
Photons        
Jets        
Btag & vertexing        
Particle Flow        
Pile-up Mike Hildreth ( Notre Dame)   %OK%  
Visualization        
Madgraph        

LONG exercises readiness

LONG Exercise name Responsible(s) Work in progress Ready (tested at FNAL) COMMENTS
longexername firstname lastname ( My University), firstname lastname ( My University) Work in progress, under construction %OK% My comments
Jets (new physics)        
Photons        
Top and top-like searches Sadia Khalil ( KSU), James Dolen ( Buffalo) Work in progress, under construction    
Exotica displaced v        
H low mass        
HZZ4l Nicola De Filippis ( Politecnico di Bari) et al. Work in progress, under construction %OK%  
Physics with taus Jeff Kolb, ( Notre Dame ), Mike Hildreth ( Notre Dame ), Keti Kaadze ( CERN) Work in progress, under construction    

Exercises Readiness for CMSDAS at Hamburg

Pre-exercises readiness

Exercise Set Responsible(s) Work in progress Ready COMMENTS
Set 1 Sudhir Malik ( UNL/FNAL)   %OK% Announced on 1 December 2012
Set 2 Sudhir Malik ( UNL/FNAL)   %OK% "
Set 3 Sudhir Malik ( UNL/FNAL)   %OK% "
Set 4 Sudhir Malik ( UNL/FNAL)   %OK% "

SHORT exercises readiness

SHORT Exercise name Responsible(s) Work in progress Ready (tested at DESY) COMMENTS
shortexername firstname lastname ( My University), firstname lastname ( My University) Work in progress, under construction %OK% My comments
Generators Altan Cakir (DESY) Work in progress, under construction    
Tracking and Primary Vertices Nazar Bartosik, Joerg Behr, Gregor Hellwig   %OK%  
Electrons Hannes Schettler (DESY) Work in progress, under construction    
Muons Ivan Asin (DESY), Carmen Diez Pardos (DESY), Francesco Costanza (DESY) Work in progress, under construction    
Photons Valentina Sola (Hamburg) Work in progress, under construction    
Jets Roman Kogler (Hamburg)   %OK%  
Btag & vertexing Roberval Walsh (DESY) Work in progress, under construction    
Particle Flow Alexei Rapsereza (DESY), Adrian Pereanu (RWTH Aachen) Work in progress, under construction    
Pile-up Dirk Kruecker (DESY) Work in progress, under construction    
Visualization (tbd) Work in progress, under construction    
ROOSTATS Jochen Ott (KIT) %ICONPwip}%    
LONG exercises readiness

LONG Exercise name Responsible(s) Work in progress Ready (tested at DESY) COMMENTS
longexername firstname lastname ( My University), firstname lastname ( My University) Work in progress, under construction %OK% My comments
Jets (new physics) Sebastian Naumann (DESY) et al. Work in progress, under construction    
Photons Christian Autermann (RWTH Aachen), Valentina Sola (Hamburg) et al. Work in progress, under construction    
Top and top-like searches Tyler Dorland (DESY) et al. Work in progress, under construction %OK%  
Exotica displaced v Alexander Schmitt (Hamburg) et al. Work in progress, under construction    
H low mass Dean Horton (DESY) et al. Work in progress, under construction    
HZZ4l Nicola De Filippis ( Politecnico di Bari) et al. Work in progress, under construction    
Physics with taus Armin Burgmeier (DESY), Alexei Raspereza (DESY), Adrian Pereanu (RWTH Aachen ) et al. Work in progress, under construction    
SUSY Matthias Schroeder (DESY) et al. Work in progress, under construction    

CMSDAS 2012 (archive) ( LPC, Pisa, Taiwan)

Before the workshop the participants are required to complete 4 Pre-Workshop Exercises. During the week of workshop, following the first day of CMS status talks, participants get a first hands-on experience on physics data analysis through Short Exercises and Long Exercises. The following topics give a summary of each of these, the names of the authors and provides a link to the twiki with the detailed exercise.

Pre-Workshop Exercises

Short Exercises

1. Roostats https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolStatisticsFall2012

Description: We offer a series of exercises on the most often needed statistics procedures in CMS physics analyses: limits, significance estimates, trials factor, hypothesis testing. The excercises attempt to keep to a rigorous theoretical treatment of the statistics issues. Above all, we focus on the practical issues of designing a statistics procedure for an analysis, choosing appropriate tools, implementing and validating the analysis. We overview RooStats as the framework of choice for statistical inference, several tools that make use of it. We discuss how to use RooStats and make your own tool. We cover the prominent methods: CLs - as a method of choice in CMS and ATLAS for limit setting, we also discuss Bayesian and frequentist approaches.

Schedule premitting, Bob Cousins will present his renowned lecture on the basics of statistical inference in particle experiment.


Responsible at CMSDASia @Taiwan(NTU): Gena Kukartsev ( Brown University), Sara Bolognesi ( JHU), Xin Shi ( NTU)
Responsible at CMSDAS@LPC: Gena Kukartsev ( Brown University), Mario Pelliccioni ( Torino), Luca Lista ( Napoli)

2. Generators SWGuideCMSDataAnalysisSchool2012GeneratorExercise SWGuideCMSDataAnalysisSchool2012GeneratorExerciseForPisa

Description:

Get to know your generators. You should leave this tutorial understanding the following concepts:

  • Different Monte Carlo tools provide predictions at different levels of approximation. You will be exposed to the terminology LO (leading order), leading log(arithm) (LL), and NLO (next-to-leading order). You will hear about these extensively in the lectures, and it is not the purpose of this tutorial to replace that. Rather, you should understand that they are different, and provide different predictions of physical observables.
  • Matched or merged calculations contain both matrix element predictions for parton emissions and parton showers.
  • Monte Carlo truth information is stored in a format defined by the HepMC standard. You should be exposed to how to navigate this record and identify, for example, an electron arising from a W- decay compared to an electron arising from a b quark decay.
  • Parton distribution functions, or PDFs, are an important input to phenomenological predictions. Many properties of events at hadron colliders are sensitive to them.
  • In Electroweak processes, such as those studied here, spin correlations can lead to important observable effects. Not all Monte Carlo tools include spin correlations, and some include them to higher order in perturbation theory than others.

Responsible at CMSDASia@Taiwan(NTU): Tai-Wei Wang ( NTU)
Responsible at CMSDAS@LPC: Steve Mrenna ( Fermilab), Fabio Cossutti ( Trieste)

3. Tracking and Primary Vertices SWGuideCMSDataAnalysisSchool2013TrackingExercise

Description:

We will present an introduction to using tracks for analyses in the era of large pile-up (many primary vertices). Our exercises will all use real data and will familiarize you with the following techniques:

  • Extracting basic track parameters and reconstructing invariant masses from tracks in CMSSW. Tracks are the closest detector entity to the four-vectors of particles: the momentum of a track is nearly the momentum of the charged particle itself.
  • Cleaning sets of tracks for analysis. We will use POG-endorsed filters to eliminate bad tracks and discuss sources of tracking uncertainties.
  • Extracting basic parameters of primary vertices. In this high-luminosity era, it is not uncommon for a single event to contain as many as ten to twenty independent collisions. For most analyses, only one is relevant, and it can usually be identified by its tracks.


Responsible at CMSDASia@Taiwan(NTU): Jim Pivarski ( Fermilab), Andrea Venturi ( Pisa), Po-Hsun Chen ( NTU)
Responsible at CMSDAS@LPC: Kevin Burkett ( Fermilab), Jim Pivarski ( Fermilab), Andrea Venturi ( Pisa)

4. Electrons SWGuideCMSDataAnalysisSchool2012ElectronExercise

Description: Detecting electrons in CMS data, from low energies to the electroweak scale and beyond. The exercise is aimed at describing a selection electrons, separating them from the background from QCD fakes or electrons from conversion. We will use an inclusive sample of triggered electrons and describe the selection variables, how to check them, and see step by step how the W and Z signals emerge from the background. At the end of the exercise, the successful student will be comfortable with arriving at their own electron selection and estimating its performance on data.


Responsible at CMSDASia@Taiwan(NTU): Yurii Maravin ( Kansas State U.), Syue-Wei Li ( NCU)
Responsible at CMSDAS@LPC: Paolo Meridiani ( INFN Rome) Bryan Dahmes ( Minnesaota)

5. Muons SWGuideCMSDataAnalysisSchool2012MuonExercise

Description: This exercise is intended to familiarize you with using muons in a physics analysis. In particular, you will learn: a brief overview of muon reconstruction and identification algorithms, what muon related information is available in AOD and PAT and how to access muon specific information that could be relevant for your analysis. We will show you what typical performance you can expect for muons in CMS, regarding efficiency, acceptance, purity and momentum resolution. The exercise will also cover the basics of measuring the muon identification and trigger efficiency in data.


Responsible at CMSDASia@Taiwan(NTU): Martijn Mulders ( CERN), Devdatta Majumdar ( NCU), Senka Djuric ( IRB), Fang-Ying Tsai ( NCU)
Responsible at CMSDAS@LPC: Adam Everett ( Purdue), Martijn Mulders ( CERN)

6. Photons SWGuideCMSDataAnalysisSchool2012PhotonShortExercise

The three short photon exercises are designed to cover the breadth of photon identification for beginners. Three basic elements of photon selection (discriminating against jets) and analysis are covered: the characteristic shower in the ECAL, the isolation in the immediate vicinity, and the measurement of efficiency (with emphasis on trigger).
Exercise 1: Photon efficiency and the HLT

  • learn the difference between HLT and off-line selection
  • learn basics of efficiency measurement in data
  • measure an efficiency of a particular HLT trigger

Exercise 2: Isolation & Multivariate analyses
  • learn the principles (jet fragmentation, underlying event, pile-up, footprint removal)
  • Particle Flow isolation in a solid cone
  • modified Frixione PF isolation using multi-variate technique and comparison to solid cone

Exercise 3: Shower shape and jet-photon mis-identification
  • learn how photons are clustered and which cluster shape variables are useful
  • learn how to extract the characteristic shapes from data and MC
  • learn how to apply shower shape variables to predict the background from jets mis-identified as photons.


Responsible at CMSDASia@Taiwan(NTU): Yurii Maravin ( Kansas State U.), Chia-Ming Kuo ( NCU), Yu-Wei Chang ( NTU)
Responsible at CMSDAS@LPC: Andrew Askew ( FSU), Yuri Gershtein ( Rutgers)

7. Jets SWGuideCMSDataAnalysisSchoolJetAnalysis

Description: The short exercise provides hands-on experience accessing jets, plotting basic jet quantities, and teaches the basics of jet energy corrections and their uncertainty. You will learn the various available pile-up corrections for jets. You will become familiar with basic jet types and algorithms, including jet sub-structure algorithms.

Responsible at CMSDASia @Taiwan(NTU): Kalanand Mishra ( Fermilab), Ilya Osipenkov ( Texas-A&M), TA: Charles Dietz ( NTU)
Responsible at CMSDAS@LPCL: Suvadeep Bose ( Nebraska), Eva Halkiakdakis ( Rutgers), Robert Harris ( Fermilab), Kalanand Mishra ( Fermilab), Salvatore Rappoccio ( Johns Hopkins), Kai Yi ( Iowa)
Responsible at CMSDAS@Pisa: Suvadeep Bose ( Nebraska), Dan Duggan ( Rutgers), Kalanand Mishra ( Fermilab)

8. Btag & vertexing https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolbTagExercise

Description: Jets that arise from bottom quark hadronisation and decay (denoted b jets) are present in a wide range of physics processes of interest, such as the decay of top quark, Higgs bosons, and various supersymmetric processes. The ability to accurately identify b jets is vital in reducing the otherwise overwhelming background to these channels from processes involving jets from gluon and u, d, s-quark fragmentation (light flavor) and from charm quark hadronisation.

In this exercise, students will learn: (1) about the properties and differences between primary vertices and the luminous region, also known as beam spot. Students will learn how to access the vertex collections and make plots of the main vertex variables.

(2) how to identify b-quark jets, plot the b-tag efficiency and mistagging rate as a function of jet pT and eta for a given tagging algorithm in MC. Use the b-tagging scale factors derived from data to correct Monte Carlo.


Responsible at CMSDASia@Taiwan(NTU): Francisco Yumiceva ( Fermilab) , Cristina Ferro ( NCU)
Responsible at CMSDAS@LPC: Francisco Yumiceva ( Fermilab) , Pratima Jindal ( University of Nebraska at Lincoln)

9. Particle Flow https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolPFlowShortExercise

Description: In this exercise, you will learn the basics of particle flow by creating your own "toy" algorithm (which you will compare with the real CMS algorithm). Particle Flow reconstruction aims at reconstructing all stable particles within an event from first principles. These particles are then used as input to reconstruct more higher-level, derived objects such as taus, jets, missing transverse momentum...just as could be done with true particles from a simulation. The basic building elements for the particle flow algorithm are tracks, electromagnetic clusters and hadronic calorimeter clusters.

Because the CMS detector is itself very sophisticated, the CMS Particle Flow algorithm is quite detailed, accounting for many subtle, intricate effects from each sub-detector. A complete exposition of the full algorithm is not possible here in a very brief two hour exercise. Indeed, the CMS Particle Flow Algorithm is well documented and detailed information can be found at SWGuideParticleFlow. Rather, the goal of this exercise is to merely introduce the very basic concepts of particle flow event reconstruction, which are quite universal and not overly experiment specific, and which are indeed both elegant and quite simple. After you finish this exercise, you should have a basic understanding of what particle flow is and how (and why) it works, particularly in the case of CMS.

Historically, the CMS particle flow algorithm was developed by first scanning events and visualizing how different types of individual particles interact in across different CMS sub-detectors. Only after a visual intuition is developed does one even attempt to implement an algorithm. We will follow that path in this simple exercise.


Responsible CMSDASia@Taiwan(NTU): Chris Silkworth ( UIC)Rick Cavanaugh ( Fermilab)
Responsible at CMSDAS@LPC: Chris Silkworth ( UIC)Rick Cavanaugh ( Fermilab)

10. Pile-up https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2012PileupReweighting

Description: Learn the procedure for reweighting MC distributions to match the pileup in the data. Students will learn how to access the various pieces of information related to pileup in the MC events, and how to use the reweighting tools.

Responsible at at CMSDASia@Taiwan(NTU): Mike Hildreth ( Notre Dame)
Responsible at CMSDAS@LPC: Mike Hildreth ( Notre Dame)

11. Visualization https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2012EventScanningExercise

Description: The goal of this short exercise is to enable the student to “scan” interesting CMS events. The tool to be used is Fireworks. This tool has been used in early cosmic ray events (CRAFT) and then in early running at 0.45 and 2.36 TeV. The representative data used here is from the period April – October 2010 when the LHC luminosity was doubling every~ 2 weeks. During that period an “Exotica Scan” was set up, which selected the ~ 30 most bizarre events each day and asked a small group of “scanners” to try to understand them. The goal was to identify problems and issues with the CMS detector.


Responsible CMSDASia@Taiwan(NTU): Yuan Chao ( NTU)
Responsible at CMSDAS@LPC: Alexey Ferapontov ( Brown University)

12. MadGraph https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2012MadGraphExercise

Description: The goal of this short exercise is to familiarize students with the automated matrix element generator MadGraph and with its functionalities for jet matching between matrix elements and parton showers. During the exercise, the students will use generate some Standard Model and SUSY processes, perform jet matching, and analyse the result. Where is matching needed? How many jets are necessary? What are the special issues with matching in BSM scenarios such as the MSSM? If time permits, also other advanced uses of MadGraph will be addressed.


Responsible CMSDASia@Taiwan(NTU): Johan Awall ( NTU)

Long Exercises

1. Jets (new physics) : SWGuideCMSDataAnalysisSchoolJetsLongExerciseDijetMassSpectrumInAssociationWithW

Description: The long exercise will teach you how to search for new physics with jets by looking at the dijet mass spectrum in association with the W boson. You will study the physics of the dijet spectrum, analyze relevant processes, learn to recognize their signatures and develop analysis techniques while examining CMS Data.

It is strongly recommended that students take the Jet Short Exercise prior to taking the Jet Long Exercise. Other related short exercises are pile-up, particle flow, and generators. Limits will not be extracted in the jet long exercise, so the Roostats short exercise is a valuable supplement to the jet long exercise for those students planning to set limits in the future.


Responsible at CMSDASia@Taiwan(NTU): Kalanand Mishra ( Fermilab), Ilya Osipenkov ( Texas-A&M)
Responsible for the long exercise with three jet resonances ( SWGuideCMSDataAnalysisSchoolJetResonances2012) at CMSDAS@LPC: Suvadeep Bose ( Nebraska), Dan Duggan ( Rutgers), Eva Halkiakdakis ( Rutgers), Robert Harris ( Fermilab), Kalanand Mishra ( Fermilab)
Responsible for the long exercise with three jet resonances ( SWGuideCMSDataAnalysisSchoolJetResonances2012) at at CMSDAS@Pisa: Suvadeep Bose ( Nebraska), Dan Duggan ( Rutgers), Kalanand Mishra ( Fermilab)

2. Photons SWGuideCMSDataAnalysisSchool2012PhotonLongExercise

There is nothing quite like doing analysis with photons to introduce new students to doing analysis with photons. Following the 2011 CMSDAS experience, we will do a physics analysis again! The long exercise will kick off an effort for the search for a three photon final state (which arises naturally from new physics with vector-like confinement)in 2011 data. It is anticipated that the major parts of the analysis will be completed at CMSDAS, and that the involved analyzers will stay involved to see the results through to publication.


Responsible at CMSDASia@Taiwan(NTU): Chia-Ming Kuo ( NCU), Yu-Wei Chang ( NTU)
Responsible at CMSDASia@LPC: Yuri Gershtein ( Rutgers), Andrew Askew ( FSU),

3. Top and top-like searches https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolTopExercise

Measurement of the top quark pair production cross section in events containing lepton+jets using b-quark identification

Description: The basic exercise measures the top quark pair production cross section using events containing one isolated muon, 3 or more jets, missing transverse energy, indicative of the presence of a neutrino, and using b-quark identification. This sample can also be used to search for new physics if there is interest in doing so.

Students will learn about the main background sources to the top quark signal in this channel, and use a data-driven method to estimate the QCD multijet background. The top quark pair production cross section will be measured from the 2011 data using two methods: template fit and counting.

Suggested short exercises: b-tagging, muons, PFlow. Essential: RooStats (fitting part)


Responsible at CMSDASia@NTU(Taiwan): Freya Blekman ( Brussels), Francisco Yumiceva( Fermilab)
Responsible at CMSDAS@LPC and CMSDAS@Pisa: Freya Blekman ( Brussels), Andrea Giammanco ( Pisa), Sadia Khalil ( KSU), Salvatore Rappoccio ( Fermilab), Francisco Yumiceva( Fermilab)

4. Exotica displaced v. https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2012ExoticaDisplacedVertices

Description: In this exercise, you will perform a search for a heavy resonance (M) decaying to two long-lived massive neutral particles (m), which then decay to leptons (electrons and muons). The full chain is M -> m m -> 4 leptons, with both pairs of leptons being displaced from the primary vertex. This highly exotic signature would be clear evidence of physics beyond the Standard Model, and is motivated by hidden valley scenarios.

You will perform the complete analysis with full statistics. The exercise will cover MC simulation, kinematics, definition of acceptance, efficiencies, and will use the RooStats tools to set upper limits (unless we see something new). We will discuss and explore sources of systematic uncertainty and how to estimate them using data-driven methods wherever possible.

The exercise will be based on 2012 data. The increased energy and statistics will lead to an increased sensitivity and provide a first look beyond the existing limits which were derived on 2011 data.

Preferred short exercises: tracking, muons, Btag & vertexing, RooStats.


Responsible at CMSDASia@NTU(Taiwan): Jim Pivarski ( Fermilab),Ian Shipsey ( Purdue), Marco De Mattia ( Purdue)
Responsible at CMSDAS@LPC and CMSDAS@Pisa: Nuno Leonardo ( Purdue), Ian Shipsey ( Purdue), Marco De Mattia ( Purdue)

5. H low mass SWGuideCMSDataAnalysisSchoolLowMassHiggsSearchExercise

Description: "The observation of a Higgs-boson like at ICHEP 2012 with a mass of 125 GeV has been an extraordinary event for CMS and for particle physics. The next goal of the experiment is now to characterize the properties of this new boson, and confirm/exclude that is behaving as the Standard Model (SM) Higgs. One fundamental property to study is the new boson coupling to fermions. We will focus on a H->bb search which has the largest decay branching ratio (~60% at 125 GeV). In this exercise we will start with an overview of the differences between signal and background processes and corresponding discriminating variables in the channel. We will try to define our selection cuts in order to maximize the probability to observe a signal in our data. Depending on time, we will try 2 different analysis scenarios (mass shape, MVA optimized) to set limits on the Higgs cross section (or observe an excess compatible with the SM Higgs). The last step: we will look into several systematic uncertainty contributions (like b-tag efficiency, jet resolution,etc.). We will check all the distributions and results with both data and MC. The methods used in the exercises will be very similar to the ones used in the official analysis."
Preferred short exercises: Jets, Particle-Flow, Btag.


Responsible at CMSDASia@NTU(Taiwan): Michele de Gruttola( UofFlorida), David Lopes Pegna( Princeton)
Responsible at CMSDAS@LPC: Michele de Gruttola( UofFlorida), Jacobo Konigsberg( UofFlorida), David Lopes Pegna( Princeton)
Responsible in CMSDAS@Pisa: Andrea Rizzi ( Pisa), Pierluigi Bortignon ( ETHZurich)

6. Higgs in 4 leptons SWGuideCMSDataAnalysisSchoolHighMassHiggsSearchExercise for LPC and Pisa SWGuideCMSDataAnalysisSchoolHiggs4LepSearchExercise for NTU

Description: The 4 leptons final state has been one of the main channel in the discovery of the boson at 125 GeV. The importance of this channel is due to the very small background and very good mass resolution. So, as first step, we will review the analysis flow to understand how we can select a clean signal. We will analyze the kinematics of the signal and the main background (ZZ) and we will extract the expected lepton efficiency and resolution. We will study the features of the reducible background (Z+jets) and we will use control region in data to cross-check the expected number of Z+jets events in the signal region. As second step, we will apply statistical analysis to the results to compute the p-value of the observed excess in data and, if time will permit, we will extract a measurement of the mass of the discovered resonance.


Responsible at CMSDAS@NTU: Sara Bolognesi ( Johns Hopkins U)
Responsible at CMSDAS@LPC: Nicola De Filippis ( INFN Bari), Sara Bolognesi ( Johns Hopkins U), Alexey Drozdetskiy ( Univ. of Florida)
Responsible at CMSDAS@Pisa: Nicola De Filippis ( INFN Bari), Marco Meneghelli ( Univ. of Bologna)

7. SUSY hadronic SWGuideCMSDataAnalysisSchool2012SUSYHadExercise

Description: In this exercise we will analyze the CMS data for any signs of new physics, such as supersymmetry. For the 2011 summer conferences the analysis to be presented produced the most stringent limits on the standard supersymmetry scenario. An essential part of the analysis is the evaluation of the standard model contributions to the jets and missing transverse energy final state. These backgrounds are extracted from the data and validated using Monte Carlo simulations. The main backgrounds originate from QCD-multijet production,W+jets and ttbar + jets and from Z->nunu+jets. The observed data event yield, the background expectations and the predictions of different signal simulations are used to calculate significances of present or absent signals. The goal of this exercise is to carry out (at least) one data-driven background prediction method and to statistically interpret the observed results.


Responsible at CMSDAS@LPC: Gheorghe Lungu ( Rockefeller), Sarah Malik ( Rockefeller), Seema Sharma ( Fermilab), Christian Auterman ( Hamburg)
This exercise is only available at LPC

8. Wprime SWGuideCMSDataAnalysisSchool2012WprimeWZSearch

Description: In this exercise we search for W' and Technicolor particles in the WZ final state, with both W and Z bosons decaying leptonically (electrons or muons). The analysis steps include the W and Z reconstruction using high-pt leptons and MET, the study and suppression of background sources and signal efficiency determination using the simulation, and the statistical analysis of the final kinematic distributions.

The goal of the exercise is to learn how to reconstruct composite particles starting from W/Z bosons, optimize the selection cuts in the search for a new heavy resonance and become familiar with the standard statistical tools used at CMS relevant for searches.

Prerequisite: "Muons", "electrons" and Roostats short exercises.


Responsible at CMSDAS@LPC: Christos Leonidopoulos ( CERN), Cory Fantasia ( Boston), Flavia Dias ( Universidade Estadual Paulista and Caltech)
This exercise is only available at LPC

9. Leptonic SUSY https://twiki.cern.ch/twiki/bin/view/CMSPublic/CMSDASJZBAnalysis

Description: This exercise is related to the study of a search for Physics beyond the Standard Model in final states with a Z boson, jets and missing transverse energy. The Jet-Z Balance method is used to estimate the total background expected in the signal region directly from the data. The two main backgrounds: Z+jets and ttbar will be estimated using control regions in the data. The size of the sample corresponds to an integrated luminosity of 2.1 fb\x{2212}1. In the absence of any significant excess beyond the SM expectation, upper limits are set in the context of simplified supersymmetric models.

Prerequisite: "Muons", "electrons", "Jets" and Roostats short exercises.


Responsible at CMSDAS@Pisa: Pablo Martinez ( ETH Zurich)
This exercise is only available at Pisa

10. Upsilon https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2012ExoticaDisplacedVertices

Description: The exercise explores data-analysis techniques applied to the measurement of low-mass dimuon signals. The upsilon family of resonances, detected at CMS as three separate Y(nS) peaks in the dimuon mass spectrum, stand as a suitable laboratory for carrying out related studies. We first explore the ingredients and techniques necessary for performing a cross section measurement at CMS, and apply these to the production of the upsilon ground state in pp collisions. Next we study the ratio of excited upsilon states relative to the ground state, in both pp and PbPb collisions. The measurement of observable ratios, relating different processes or datasets, involve the cancellation of various terms and related systematic uncertainties, and illustrate the benefits of such approaches in analysis design. The double-ratio (excited vs ground states, PbPb vs pp datasets) is employed to search for the disapearance of the excited, least-bound states, expected in the presence of quark gluon plasma formation.

This search has been published by CMS based on the 2010 datset, and yielded the first positive evidence of upsilon suppression, with a statistical significance of 2.4 sigma. The 2011 dataset, currently being collected with up to a 20 times larger size, will be explored aiming at a first definitive observation.

The exercise involves various general techniques, including data-based tag-and-probe and statistical methods, of relevance for both measurements and searches. It is also aimed at introducing concepts and techniques and emphasizing synergies accross the pp and heavy-ion environments, where students with main interests in each of the two areas are expected to participate.


Responsible at CMSDAS@LPC: Nuno Leonardo ( Purdue), Ian Shipsey ( Purdue), Marco De Mattia ( Purdue)
This exercise is only available at LPC

11. Zprime SWGuideCMSDataAnalysisSchool2012DimuonExercise

Description: This extended exercise is intended to familiarize you with two dimuon analysis efforts in CMS. The Drell-Yan to dimuon spectrum is a fruitful signature for discovering new physics, and we seek to measure the Drell-Yan spectrum to high precision in CMS. Many new models of physics predict a new narrow resonance, here referred to generically as Z', that decays into a pair of oppositely-charged muons. We look for a bump from Z' in the dimuon invariant mass spectrum using a shape-based search, looking for the difference in between the exponentially falling background and the resonance peak that would show up should a Z' exist. In this exercise, you'll work on the meat of the analysis to produce the dimuon mass spectra for data and MC.

In doing this exercise you can learn (among other things): how to apply some loose and tight selection criteria to select muons and make combined dimuon object collections; how to evaluate acceptance, trigger efficiency, mass resolution; how to estimate some of the backgrounds using the data itself.

Prerequisite: "Muons" short exercise. Additional highly recommended prerequisite: "Particle Flow", "Tracking"


Responsible at CMSDAS@LPC: Adam Everett ( Purdue)
This exercise is only available at LPC

12. Physics with taus SWGuideCMSDataAnalysisSchoolPhysicsWithTausExercise

Description: "Taus are priviliged probes of new physics. In the context of the CMS Higgs boson(s) searches, final states with taus are well established and promising channels. With the whole of 2011 data set, the Higgs to tau-tau channel alone can set a limit on the Higgs boson cross-section from 3 to 5 times the SM prediction in the range 115-145 GeV. This has been achieved thanks to the combination of multiple tau final states and by exploiting the different production modes (gg-fusion, vector boson fusion) as to maximize the statistical power of the full data set. In models with extended Higgs sectors, like the MSSM, the coupling of the Higgs boson(s) to taus can be significantly enhanced depending on the model parameters: this makes the tau lepton a probe for supersymmetry. In this exercise, a simple cut-based analysis is proposed to isolate a clean sample of events where a Z boson decays to a pair of taus, which in turn decay to a muon (plus neutrinos) and pions (plus neutrino). The students will be asked to produce the visible mass spectrum of the tau-pair, which noticeably exhibits a clean Z peak. The basic steps of a search analysis will be presented: the SM backgrounds will be calibrated in control regions pretty much as done in the official CMS analysis; a set of data/MC correction factors will be proposed to correct the MC efficiency. Depending on the time, the results of the analysis will be used as a simple counting-experiment to set an upper limit on the SM Higgs boson cross-section."
Preferred short exercises: Particle-Flow, Muon.


Responsible at CMSDASia@NTU(Taiwan): Mike Hildreth( Notre Dame)
Responsible at CMSDAS@Pisa: Lorenzo Bianchini (Ecole Polytechnique), Simone Gennai ( CERN, INFN-Milano); Taipei Facilitator: Mike Hildreth (Notre Dame)

12. SM EWK XS measurement SWGuideCMSDataAnalysisSchoolSMEwkCrossSectionMeasurement

Description:"This exercise is related to the study of Zgamma production in electron channel. The major background is Z+Jets which Jets are are identified as photons. The students will learn a data-driven method to estimate this background."

Preferred short exercises: Electron, Photon.


Responsible at CMSDASia@NTU(Taiwan):Syue-Wei Li( NCU)

Readiness - SHORT and LONG exercises for CMSDASia Taiwan

Pre-exercises readiness

Exercise Set Responsible(s) Work in progress Ready COMMENTS
Set 1 Sudhir Malik ( UNL/FNAL), Nitish Dhingra ( Punjab U), Yu Zheng ( Purdue)   %OK% Announced on 2 Aug 2012
Set 2 Sudhir Malik ( UNL/FNAL), Nitish Dhingra ( Punjab U), Yu Zheng ( Purdue)   %OK% "
Set 3 Sudhir Malik ( UNL/FNAL), Nitish Dhingra ( Punjab U), Yu Zheng ( Purdue)   %OK% "
Set 4 Sudhir Malik ( UNL/FNAL), Nitish Dhingra ( Punjab U), Yu Zheng ( Purdue)   %OK% "

SHORT exercises readiness

SHORT Exercise name Responsible(s) Work in progress Ready (tested at Taiwan) COMMENTS
Example Name(s) Work in progress, under construction %OK% my comments
Roostats Gena Kukartsev ( Brown University), Sara Bolognesi ( JHU), Xin Shi ( NTU) Work in progress, under construction    
Generators Tai-Wei Wang ( NTU) Work in progress, under construction
Tracking and Primary Vertices Jim Pivarski ( Fermilab), Andrea Venturi ( Pisa), Po-Hsun Chen ( NTU) Work in progress, under construction    
Electrons Yurii Maravin ( Kansas State U.), Syue-Wei Li ( NCU) Work in progress, under construction    
Muons Martijn Mulders ( CERN), Devdatta Majumdar ( NCU), Senka Djuric ( IRB), Fang-Ying Tsai ( NCU) Work in progress, under construction    
Photons Yurii Maravin ( Kansas State U.), Chia-Ming Kuo ( NCU), Yu-Wei Chang ( NTU) Work in progress, under construction    
Jets Kalanand Mishra ( Fermilab), Ilya Osipenkov (T_exas A&M_), TA: Charles Dietz ( NTU)   %OK% Announced on 14 Aug 2012
Btag & vertexing Francisco Yumiceva ( Fermilab) , Cristina Ferro ( NCU)   %OK%  
Particle Flow Chris Silkworth ( UIC)Rick Cavanaugh ( Fermilab) Work in progress, under construction    
Pile-up Mike Hildreth ( Notre Dame)   %OK%  
Visualization Yuan Chao   %OK%  
Madgraph Johan Alwall ( NTU) Work in progress, under construction    

LONG execises readiness

LONG Exercise name Responsible(s) Preferred short exercises Work in progress Ready (tested at Taiwan) COMMENTS
Example Name(s) Preferred short exercises Work in progress, under construction %OK% my comments
Jets (new physics) Kalanand Mishra ( Fermilab), Ilya Osipenkov ( Texas A&M) Essential: Jets, Roostats. Also recommended: Generators Work in progress, under construction    
Photons Chia-Ming Kuo ( NCU), Yu-Wei Chang ( NTU)   Work in progress, under construction    
Top and top-like searches Freya Blekman ( Brussels), Francisco Yumiceva( Fermilab) b-tagging, muons, PFlow. Essential: RooStats (fitting part) Work in progress, under construction    
Exotica displaced v Jim Pivarski ( Fermilab), Ian Shipsey ( Purdue), Marco De Mattia ( Purdue) tracking, muons, Btag & vertexing Work in progress, under construction    
H low mass Michele de Gruttola( U of Florida), David Lopez Pegna( Princeton) Jets, Particle-Flow, Btag & vertexing Work in progress, under construction    
H high mass Sara Bolognesi ( JHU) Electrons, Muons Work in progress, under construction    
Physics with taus Mike Hildreth( Notre Dame) Particle-Flow, Muon Work in progress, under construction  
SM EWK XS measurement Syue-Wei Li( NCU) Electron, Photon Work in progress, under construction    

CMSDAS 2011 (archive) (LPC)

Short Exercises CMSDAS 2011

1. Jets: SWGuideCMSDataAnalysisSchoolJetAnalysis


Description: Give you a hands-on experience on how to access jet collection in an event, plot basic jet quantities, and apply jet energy correction.
  • A 101 on how to access jets in the CMS framework without assuming prior knowledge of jet analysis.
  • Make you familiar with basic jet types and algorithms and how to use them in your analysis.
  • Illustrate each exercise using real life example scripts.
  • Give you a comprehensive reference to more advanced workbook examples, additional resources, and pedagogical documentation in one place.

Responsible:John Paul Chou ( Brown), Jason St. John ( Boston), Kalanand Mishra ( Fermilab), Robert Harris ( Fermilab), Eva Halkiadakis ( Rutgers) 2. Generators - Monte Carlo Tools: SWGuideCMSDataAnalysisSchoolGeneratorExercise

Description: You should leave this tutorial understanding the following concepts:
  • Different Monte Carlo tools provide predictions at different levels of approximation. You will be exposed to the terminology LO (leading order), leading log(arithm) (LL), and NLO (next-to-leading order). You will hear about these extensively in the lectures, and it is not the purpose of this tutorial to replace that. Rather, you should understand that they are different, and provide different predictions of physical observables.
  • Matched or merged calculations contain both matrix element predictions for parton emissions and parton showers.
  • Monte Carlo truth information is stored in a format defined by the HepMC standard. You should be exposed to how to navigate this record and identify, for example, an electron arising from a W- decay compared to an electron arising from a b quark decay.
  • Parton distribution functions, or PDFs, are an important input to phenomenological predictions. Many properties of events at hadron colliders are sensitive to them.
  • In Electroweak processes, such as those studied here, spin correlations can lead to important observable effects. Not all Monte Carlo tools include spin correlations, and some include them to higher order in perturbation theory than others.
  • ROOT is a tool for packing the HepMC record and for visualizing it. While the basic tutorial does not rely on any individual knowledge of ROOT, there are ample opportunities for students to learn how to write C++ code to work with ROOT.

Responsible: Stephen Mrenna ( Fermilab), Charles Plager ( UCLA/Fermilab)

3. ExoStive RooSting: SWGuideCMSDataAnalysisSchoolExoStiveRooStingExercise


Description: Basics of data modeling with RooFit and interval finding with RooStats. Configuration file interface to RooFit and RooStats (ExoSt).

Responsible: Gena Kukartsev ( Brown), Petar Maksimovic ( Johns Hopkins)

4. BTag and Vertexing: SWGuideCMSDataAnalysisSchoolbTagExercise

Description: Learn about the properties and differences between primary vertices and the luminous region or as also known as beam spot. Students will learn how to access the vertex collections and make plots of the main variables.Learn how to identify b-flavored jets and plot the b-tag efficiency and mistagging rate as a function of jet kinematics for a given algorithm. Use the b-tagging scale factors derived from data to correct the Monte Carlo efficiencies.

Responsible: Meenakshi Narain ( Brown), Francisco Yumiceva ( Fermilab), Samvel Khalatyan ( UIC)

5. Tracking: Link to part 1 of the exercise Link to part 2 of the exercise

Description: The main goal of the tracking exercise is to give you a basic introduction to accessing and using the reconstructed tracks and vertices in an event. In this exercise, the user will learn:

  • how to access the collections of tracks and vertices in the event
  • how to search for tracking variables relevant for your analysis ("shopping for data")
  • how to work with the track geometry: calculating a vertex
  • how to build invariant mass distributions and search for resonances.

Responsible: Kevin Burkett ( Fermilab), Jim Pivarski ( Texas A&M)

6. Photons: SWGuideCMSDataAnalysisSchoolPhotonShortExercise

Description: Photons are a useful analysis object, not just for standard model measurements, but also for searches for new physics. But unlike electrons, photons have only their isolation and shower shape within the electromagnetic calorimeter to rely upon. These three short exercises will introduce newcomers to the challenges of identifying photons and measuring their efficiency.

Responsible: Andrew Askew ( Florida State University), Vasundhara Chetluru ( Fermilab)

7. Electrons: SWGuideCMSDataAnalysisSchoolElectronShortExercise

Description: Electrons are very important to the success of CMS. Several Standard Model signatures (W, Z, and top are good examples) and evidence for New Physics (SUSY, Z', etc.) can be observed by reconstructing decays that involve electrons and/or muons.In many cases, a decay involving electrons is rare. Taking Z decay as an example, only about 6% of Z bosons decay to a pair of electrons or muons (electrons make up half of this fraction) while the dominant decay process involve hadrons. But the rare occurence of electrons makes them interesting. LHC collisions are dominated by QCD processes, so the bulk of the data is full of hadronic activity (i.e. jets). If we exploit some of the features of good electrons, we can isolate a sample of interesting events from the bulk of the data. In this exercise, we will examine electron reconstruction using a portion of the 2010 dataset.

Responsible: Jeff Berryhill ( Fermilab), Bryan Dahmes ( University of Minnesota), Paolo Meridiani ( ETH Zurich)

8. Particle Flow: SWGuideCMSDataAnalysisSchoolPFlowShortExercise

Description: In this exercise, you will learn the basics of particle flow by creating your own "toy" algorithm (which you will compare with the real CMS algorithm). Particle Flow reconstruction aims at reconstructing all stable particles within an event from first principles. These particles are then used as input to reconstruct more higher-level, derived objects such as taus, jets, missing transverse momentum...just as could be done with true particles from a simulation. The basic building elements for the particle flow algorithm are tracks, electromagnetic clusters and hadronic calorimeter clusters.

Because the CMS detector is itself very sophisticated, the CMS Particle Flow algorithm is quite detailed, accounting for many subtle, intricate effects from each sub-detector. A complete exposition of the full algorithm is not possible here in a very brief two hour exercise. Indeed, the CMS Particle Flow Algorithm is well documented and detailed information can be found at SWGuideParticleFlow. Rather, the goal of this exercise is to merely introduce the very basic concepts of particle flow event reconstruction, which are quite universal and not overly experiment specific, and which are indeed both elegant and quite simple. After you finish this exercise, you should have a basic understanding of what particle flow is and how (and why) it works, particularly in the case of CMS.

Historically, the CMS particle flow algorithm was developed by first scanning events and visualizing how different types of individual particles interact in across different CMS sub-detectors. Only after a visual intuition is developed does one even attempt to implement an algorithm. We will follow that path in this simple exercise.


Responsible: Rick Cavaunagh ( UIC), Christopher Silkworth ( UIC)

9. Muons: SWGuideCMSDataAnalysisSchoolMuonExercise

Description: This exercise is intended to familiarize you with using muons in a physics analysis. In particular, you will learn:

  • how to access basic muon quantities
  • how to compare data and Monte Carlo distributions for muons
  • how to access muon specific information that could be relevant for your analysis o global muon quality o tracker muon quality o cosmic muon compatibility o shower information o different muon track fits * how to study cuts for muon object selection and event selection

Responsible:

10. Event Display: SWGuideCMSDataAnalysisSchoolEventScanningExercise

Description: The goal of this short exercise is to enable the student to “scan” interesting CMS events. The tool to be used is Fireworks. This tool has been used in early cosmic ray events (CRAFT) and then in early running at 0.45 and 2.36 TeV. The representative data used here is from the period April – October 2010 when the LHC luminosity was doubling every~ 2 weeks. During that period an “Exotica Scan” was set up, which selected the ~ 30 most bizarre events each day and asked a small group of “scanners” to try to understand them. The goal was to identify problems and issues with the CMS detector.

Responsible: Dan Green ( Fermilab), Sudhir Malik( UNL/Fermilab)

11. MET and High-MET Event Scan Exercise: SWGuideCMSDataAnalysisSchoolMETExercise

Description: Missing transverse energy (MET) is one of the most important observables used to observe basic Standard Model (SM) processes and search for new physics beyond the Standard Model. The first part of this exercise will introduce CMS newcomers to different MET reconstruction algorithms available in CMS and will provide hands-on experience accessing different types of MET objects using CMSSW (both within and outside the PAT framework). The second part will be devoted to a scan of high-MET events from the 2010 collision data where events are selected and inspected using the existing CMSSW tools.

Responsible: Jordan Damgov ( Texas Tech University), Dinko Ferencek ( University of Maryland)

12. OSET Short Exercise: SWGuideCMSDataAnalysisSchoolOSETExercise

Description:The goal of the exercise is to demonstrate how to process one Monte Carlo sample to generate another one using selection and reweighting techniques. We will investigate:
  1. How to identify a particle in a Monte Carlo truth event record.
  2. How to determine the probability that something happened.
  3. How to reweight based on a different probability.

Responsible: Stephen Mrenna ( Fermilab), Charles Plager ( UCLA/Fermilab)

Long Exercises CMSDAS 2011

1. Search for dijet Resonances: https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolJetResonances

Description: The purpose of this long exercise is to build a foundation for performing analyses at a hadron collider detector. We will do this by focusing on the ongoing search at CMS for dijet resonances.The final goal is to teach students how to do a simple search for a dijet resonance signal. The basic physics of dijet resonance signals and QCD background will be taught, and the analysis techniques used to analyze the data will be explored. In addition to the dijet resonance search, we will also look at the dijet forward-central ratio. We will use the technical experience gained in the pre-exercises and short-exercises a a basis for the long exercise. But unlike the earlier exercises, we will emphasize analysis techniques rather than strictly the software machinery. Of course, we will continue to develop and hone the machinery as we go along, but we will treat CMS software and ROOT as a means rather than an end. The course design is intended for starting graduate students with little or no prior experience in high energy physics analysis techniques.

Responsible: John Paul Chou ( Brown), Jason St. John ( Boston), Kalanand Mishra ( Fermilab), Robert Harris ( Fermilab), Eva Halkiadakis ( Rutgers)

2. Search for the Z prime: https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolZPrimeExercise

Description: This extended exercise is intended to familiarize you with the Z'-to-dimuon analysis effort in CMS. For background reading: the twiki page ExoticaZprimeMumu, the full documentation of the dimuon analysis in AN-10-317, and the paper currently in CWR, EXO-10-013, which reports the results of the combined analysis together with dielectrons. In doing this exercise you can learn (among other things):
  • how to create Z'-to-dimuon-oriented PAT tuples having needed/useful quantities embedded
  • how to select datasets and determine which runs/lumisections to use
  • how to evaluate acceptance, trigger efficiency, mass resolution
  • how to choose cut values for event and object selection (e.g. muon isolation)
  • how to estimate some of the backgrounds using the data itself.

Responsible: Adam Everett ( Purdue), Jordan Tucker ( UCLA)

3. Measurement of the Upsilon cross section: SWGuideCMSDataAnalysisSchoolUpsilonLongExercise

Description:In these Exercises we will measure the Upsilon resonance cross section as presented in arXiv:1012.5545 ( CMS-AN, approval talk). We will reproduce the same results as obtained in the paper using the 3/pb real CMS data. We will learn how to carry out a complete analysis from soup to nuts.

The two principal motivations for the study of the inclusive differential production cross section of the \x{03a5}(nS) family of resonances at CMS are: (i) elucidation of the physical processes (hadroproduction) that produce the \x{03a5}(nS) family in proton-proton collisions, and (ii) calibration of the CMS detector for low pT muons.

The differential cross section is given by %BEGINLATEX% \begin{displaymath} \frac{d\sigma (pp \to \Upsilon(nS))}{dp_T} \times Br(\Upsilon(nS) \to \mu^{+}\mu^{-}) = \frac{N^{fit}_{\Upsilon(nS)}(p_T; {\cal{A}}, \epsilon)}{\int {\cal{L}} dt \cdot \Delta p_T} \end{displaymath} %ENDLATEX%

The determination of the l.h.s is obtained from the r.h.s measured inputs:

  • fitted yield N
  • acceptance A
  • efficiency \x{03b5}
  • and luminosity L.
The implementation presented is based on the analysis of dimuon events. Online selection is based on the detection of two muons at the hardware trigger level, without an explicit pT requirement and without further selection at the HLT. Such criteria, forming the L1 trigger path, was sufficient to maintain a dimuon trigger without prescaling at the instantaneous luminosities of the LHC start-up. All three muon systems, DT, CSC and RPC, take part in the trigger decision.

The basic objects, referred to as tracker muons, correspond to tracks reconstructed in the silicon tracker and associated with a compatible signal in the muon detectors. (Read more about tracker muons, global muons, standalone muons, and Calo muons)

Along with executing the various steps involved in a physics analysis, the attendee will have become familiar with:

  • determining detector acceptances
  • data based methods for measuring reconstruction efficiencies
  • mc generation and sample manipulations, reconstruction, fitting
  • evaluation of systematic uncertainties
  • using various standard CMS software tools and packages.

Responsible: Jacob Anderson ( Fermilab), Zoltan Gecse ( Purdue), Zhen Hu ( Purdue), Nuno Leonardo ( Purdue), Ian Shipsey ( Purdue), YuZheng ( Purdue)

4. Top pair production Cross Section Measurement: SWGuideCMSDataAnalysisSchoolTopExercise

Description: Measure the top pair production cross section in lepton+jets+MET by simultaneously fitting a sample with 3 jets and 4 or more jets. Students will learn about the main background sources as a function of number of jets, the estimation the QCD multijet background using a data-driven method, and extraction of the top pair signal.

Responsible: Salvatore Rappoccio ( Johns Hopkins), Francisco Yumiceva( Fermilab), Sadia Khalil ( Kansas State), Cecilia Gerber ( UIC)

5. Search for the Higgs (high mass): SWGuideCMSDataAnalysisSchoolHiggsSearchExercise

Description: If LHC runs according to expectations 2011 wil be a great year, first for limits and then for possible discoveries. There are several "high mass Higgs" channels. The main contributor into setting limits is H -> ZZ -> 2l 2nu.

In this exercise we will start with an overview of the differences between signal and background processes and corresponding discriminating variables for the channel. We will continue with up to 3 different analysis scenarios (robust, optimized and MVA optimized) to set limits on the Higgs cross section. As last step we will look into several systematic uncertainty contributions (like momentum scale for leptons, JetMET uncertainties, etc) including their evaluation and effect on the final result. Given enough time we will overview and discuss the full list of systematic uncertainties and update accordingly the final results. All the distributions and results will be checked with both data and MC.


Responsible: Alexey Drozdetskiy ( Univ. of Florida), Andrey Korytov ( Univ. of Florida), Jonatan Piedra ( Universidad de Cantabria)

6. Photons - New Phenomena: Link to the exercise

Description: Searching for new phenomena with photons is a challenging business. But many signs of physics beyond the standard model manifest themselves with photons in the final state (not to mentions certain standard model predictions cough*Higgs*cough). In this exercise we will embark on a completely new analysis, a search for vector-like confinement in the three photon final state. At the end of CMSDAS, we should have the lion's share of a brand new paper ready to move along to publication.

Responsible: Andrew Askew ( Florida State University), Yuri Gershtein ( Rutgers University)

7. Search for Black Holes production in 2010 collision data: SWGuideCMSDataAnalysisSchoolBlackHolesLongExercise

Description: In this exercise participants will conduct a search for microscopic black hole production and decay in p p collisions at a center-of-mass energy of 7 TeV using a data sample corresponding to an integrated luminosity of 35 pb\x{2212}1 . Events with large total transverse energy will be analyzed for the presence of multiple high-energy jets, leptons, and photons, typical of a signal expected from a microscopic black hole. We will start with producing our own data, background, and signal samples. We will learn how to plot various distributions, find a good variable that discriminates between signal and background, and use this variable to develop a novel data-driven method of background estimation. Using a number of signal samples, we will optimize the offline selection criteria to increase the search sensitivity. Participants will be asked to produce a couple of event displays with signatures expected from a black hole. Further, we will set limits on the black holes production at the LHC. If time allows, we will estimate systematic uncertainties on signal.

Responsible: Alexey Ferapontov( Brown), Greg Landsberg( Brown), Patrick Tsang ( Brown)

-- SudhirMalik - 17-Nov-2011

Edit | Attach | Watch | Print version | History: r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r1 - 2016-03-11 - eduardo
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback