Introduction

This twiki page is mainly for physics validation of JetMet RelVal. It will explain plots and data types and if necessary, some technique will be included.

Release monitoring

In most cases, one can get the the release information in the hypernews RelVal. Then, from the hypernews, there are some validation details, for example, what the reference and target release are, what's the difference between reference and target and so on which are useful for the following validation check. In the bottom of this hypernews page, some helpful links for instructions and release notes are provided. Go into this link relmon and find the release comparison that you need to do to start the validation. In General, the FullSimReport, FullSimReport_PU are necessary for MC validation, which one to choose depends on your task.

When in the relmon page of one release comparison, go the section Summary Table to select samples in the first row for JetMET group which can be found in the first column.


SummaryTable.png

For JetMET RelVal validation, the RelValQCD_FlatPt, RelValTTbar, RelValZEE and RelValZMM need to be check one by one. When find the right sample and JetMET group, click the round icon go into the JetMET summary page as following as an example. There are some sub-directories including MET, Jet, METValidation and so on. For this twiki, we mainly explain plots for sub-directory METValidation.


JetMETSummary_1.pngJetMETSummary_2.png

Start Validation

Choose the METValidation sub-directory, it also contains some different sub-directories for different types of MET.


METVal_1.pngMETVal_2.png

There are

In some case, some reports of some samples may lack of one or more sub-directories listed above. Here, we mainly explain plots for sub-directory pfMet, the plots types in different sub-directories are similar. Click the icon of pfMet, in this page, there are some comparison plots for reference(black) and target(blue), we need to check the failing comparison which means that some variables in the reference release and target release have discrepancies. In the following, the plots of different variables will be explained.
  • MET : The Missing ET distribution. We expect that the distribution should be a fast falling smooth distribution for all MET types (Calo, PF)
  • MET_Phi : The Missing ET Phi. The distribution should be smooth without any spikes for PFMet and CaloMET and it is sensitive to many detector effects. Thus it is not completely flat, but a certain modulation can be observed. A big shift of the shape itself can point to issues as well. For 25ns runs a peak at 1 can be observed. This is a feature of the particle flow reconstruction in HF. This minor issue affects only very low MET Values and METPhi, so if it happens, the run is still good.
  • MEx : The Missing ET x component. We expect this to be smooth with no spikes for all MET types (Calo, PF).
  • MEy : The Missing ET y component. We expect this to be smooth with no spikes for all MET types (Calo, PF).
  • METResolution_XXX :
  • METSig :
  • METDeltaPhi_GenMETTrue :
  • METFine :
  • METDifference_GenMETXXX:
  • MET_Nvtx :
  • HFEMEt :
  • *

-- YingAn - 2020-05-13

Topic attachments
I Attachment History Action Size Date Who Comment
PNGpng JetMETSummary_1.png r1 manage 161.0 K 2020-05-13 - 13:53 YingAn  
PNGpng JetMETSummary_2.png r1 manage 209.1 K 2020-05-13 - 13:53 YingAn  
PNGpng METVal_1.png r1 manage 173.4 K 2020-05-13 - 14:06 YingAn  
PNGpng METVal_2.png r1 manage 204.5 K 2020-05-13 - 14:06 YingAn  
PNGpng SummaryTable.png r1 manage 218.3 K 2020-05-13 - 13:39 YingAn  
Edit | Attach | Watch | Print version | History: r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r1 - 2020-05-13 - YingAn
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Sandbox All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright &© 2008-2023 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback