Photon Production in Association with Jets

Documentation

Document Date Contact
TS2011_035 August 2011 M. Anderson
AN-11-128 July 2011 M. Anderson

Introduction

Outline

Goal: to measure distribution of jet multiplicity.

General steps:

  1. Run analyzer to create TTrees containing information about lead photon and jets
  2. Fit a distribution to determine the amount of signal in the data (signal extraction)
  3. Apply efficiency corrections (photon ID efficiency is a function of number of jets)
  4. Finally, unfold the jet multiplicity distribution

Running the TTree maker

Check out and compile the photon + jets analyzer:

cmsrel CMSSW_4_2_7_patch2
cd CMSSW_4_2_7_patch2/src/
cmsenv
cvs co -d PhotonJetAnalysis/PhotonsFromWenu/ UserCode/MBAnderson/PhotonJetAnalysis/PhotonsFromWenu/
scramv1 b

Edit the file photonjet_cfg.py and point it to a few AOD root files (for testing), and to run it:

cd PhotonJetAnalysis/PhotonsFromWenu/test/
cmsRun photonjet_cfg.py >& log &

Signal extraction

The file scripts/find_num_sig.py can be used to do binned likelihood fits to determine the amount of signal and background in data.

It requires three root files with TTrees in them: one for data, one for background and one with signal. Set these variables to the file names: file_data, file_signal, file_backgd.

A fit is performed for each directorly listed in the dir variable, for example:

dirs = ["g_pass_Liso_barEta_0jets30","g_pass_Liso_barEta_1jets30",
        "g_pass_Liso_barEta_2jets30","g_pass_Liso_barEta_3jets30",
        "g_pass_Liso_endEta_0jets30","g_pass_Liso_endEta_1jets30",
        "g_pass_Liso_endEta_2jets30","g_pass_Liso_endEta_3jets30"
        ]

Within each of those directories needs to be a histogram with a specific name set by name_templateHist. That histogram is grabbed from the three files, and used to do the fits.

To run the script, simply type:

./find_num_sig.py

It outputs the signal faction to the screen, and it will also create a file set by the tfile_data_scaled_summed variable (currently "Hists_Data_scaled_summed.root"), which contains data scaled by the amount of signal found by the fits.

Correct for Efficiency

The output from the previous step is the jet multiplicity distributions, scaled by the amount of signal from the fits.

To correct for efficiency, simply create a histogram that contains efficiency as a function of jet multiplicity. Simply divide the jet multiplicity distribution by the efficiency distribution (so they both need to have the same x-axis of jet multiplicity).

Correcting for efficiency is the easy part. Determining the efficiency is the hard part. TagAndProbe was used, but there may be enough data to use photons from Zgamma events to more accurately determine the efficiency.

Unfold

You may unfold using whatever software or scripts you are comfortable with. RooUnfold was used originally. An example of how to use it is in scripts/RooUnfoldExample.cxx. You must download and compile RooUnfold, and place that script within its main directory and run it like so:

root -l
.L RooUnfoldExample.cxx
RooUnfoldExample()

-- MikeAnderson - 17-Aug-2011

Edit | Attach | Watch | Print version | History: r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r2 - 2011-08-17 - MikeAnderson
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Main All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback