Recent modifications in jetid_eff package

• v03-00-04 February 17, 2009: prelim SPC for vtx confirmation only
• v03-00-02 November 8, 2008: We have rederived the Winter2009 JetID spc files of Data/MC JetID eff scale factor on the full Win09 dataset
  Winter 2009 instructions are unchanged w.r.t. what we wrote in the previous email of Nov 6th as below; if still upgrading from previous p20 release, please read the config example or see also updated instructions on the JetID web page
  .
• v02-00-05 June 7, 2008:
  There has been some confusion on what jetid_eff tag and what spc files subdirectory to use for a p20 analysis. We observe a stable behaviour of the Data/MC JetID SF in p20 pre- and post-shutdown data. So far only separate spc files for these two periods had beed provided. This is of course clumsy. In vjets_cafe, the transition to p20 pres-hutdown has been already made and may be kept for the moment. Due to larger statistics of events used for JetID studies for p20 pre-shutdown data (3.2M) compared to about 0.6M in the post-shutdown period, it is statistically safer to continue using p20 preshutdown for analyses running over the whole p20 data.
  Current latest tag is jetid_eff v02-00-05 where I added subdirectory jetid_eff/jetid/p20_all/. This directory is meant to be used by analysers running over full p20 data. For the moment, this just contains p20 preshutdown spc files, which contain most of the stats.
  Only one change was made: the trigger version in was changed by hand in the spc files in jetid_eff/jetid/p20_all/ from v15 to v15-v16 just for cosmetics and consistency reasons.
  As soon as I finish running over p20 pre-shutdown together with post-shutdown up to (including) Ichep2008, the spc files will be updated and tagged again in the same jetid_eff/jetid/p20_all/ directory.
  The above tag also contains updates to trigger eff. jetid_eff/qcdjets/ spc files by Kayle (since v02-00-04).
  So, what to use in the config for the JetCorr processor for running over the full p20 dataset?
  
  jet_correction.EffName: MCJetIDScale_finedeta
# if using after JSSR:
# jet_correction.EffName: MCJetIDScaleSSR_finedeta
jet_correction.Path: jetid_eff/jetid/p20_all/
jet_correction.TriggerVersion: v15-v16
or explicitly (and equivalently for the moment) the p20 pre-shutdown:
  
  jet_correction.EffName: MCJetIDScale_finedeta
# if using after JSSR:
# jet_correction.EffName: MCJetIDScaleSSR_finedeta
jet_correction.Path: jetid_eff/jetid/p20_pre/
jet_correction.TriggerVersion: v15

  
  I am currently running over the full p20 dataset, but it may not be available for users in time for summer results. Still, current spc files are stable and reliable enough. Let me know about questions you may have. Cheers and apologies for the confusion and too long email I doubt anyone has read till this point,
  Yours cheerfully,
  Jiri
  See also June 6th Friday's convener's meeting JetID update for more details.
• v02-00-03 May 30, 2008: JetID Data, MC, as well as Data/MC SF efficiencies for the Ichep2008 data set.
• v02-00-00 Apr 22, 2008: First p20 Nov2007-pre-shutdown JetID efficiencies are ready to test and use; these include both individual Data and MC efficiencies (mostly for the use of the QCD cross sections measurements) as well as Data/MC JetID scale factor (for most of physics groups; to correct for Data/MC JetID eff. differences). As in p17, threshold jet reconstruction efficiencies are taken care by JSSR (threshold removal).
  The appropriate package tags are
  p20: (new) jetid_eff v02-00-00
  If you want to make sure you stay with p17, use previous jetid_eff v01-01-06
  Recommended spc files to be used are those with finer eta binning, see files in jetid_eff/jetid. The Data/MC SF was derived using MC without as well as with full p20 JSSR as in caf_mc_util p21-br-62 (i.e. full shift+smear+removal in all eta regions).
  Spc files are used by caf_eff_utils::JetCorr processor to get weights, upon which jets are randomly discarded from MC by the caf_mc_util::JetRemoval processor. For example, V+Jets master config should use (after JSSR) MCJetIDScaleSSR_finedeta_Jet_Standard_2D_deta_pt_Binned.spc just as before via
  jet_correction.Path: jetid_eff/jetid/
  jet_correction.EffName: MCJetIDScaleSSR_finedeta
  only by updating to the new jetid_eff package tag.
  Efficiencies, method, results, and basic usage are documented in d0note 5634.
  
• v01-00-03 : Checking in the p21 turn-ons from Kayle. This addition includes separate parametrizations for L1, L2, L3 in addition to combined turn-ons. Trigger matching at L1/L2/L3 has been used for all of these turn-ons. The pt is in uncorrected jet pT.
• v01-00-02 : Adding single jet triggers for RunIIb from Kayle DeVaughan.
• v01-00-01 : Replacing the outdated and non-standard parameterized JetID efficiency files with new and more standard binned JetID efficiencies. Coarse and fine binning are provided; fine binning is recommended for users. Scale factors with and without SSR are provided although differences are small.
• v01-00-00 : from Thomas Gadfort :
  I've made a new tag of the jetid_eff package named v01-00-00. In this package you will find a directory called p17.09.03 with new turn on curves for each condition at L1, L2, and L3. I'm sure I got all of them for the mu+jets analysis, but I'm not sure if there are more conditions for the e+jets or alljets analyzes.
  If you doing one of these analyzes and you want p17 turn on curves, please check out the new package and let me know if I missed something. The new curves are for each trigger level and not for each trigger, which is a different strategy from before. The mu+jets caf_trigger processor has been modified to work with these new turn on curves.
  If you want to test the new curves, do the following.
  1). Get the new mu+jets config file from the head version of top_cafe. Or get it here /work/greenlake-clued0/tgadfort/cafe_cvs/top_cafe/configs/trigger/caf_trigger_mujets-base.config
  2). Get the new jetid_eff package. v01-00-00
  3). Get the new caf_trigger package. v02-04-14
  4). Run over some MC using the new packages.
  This only works for mu+jets right now. The e+jets processor will probably have to be modified.