Skip to content

DerivationFrameworkHiggs - Refine quadruplet selection for vertex fit

Johannes Junggeburth requested to merge jojungge/athena:FourLep_Vtx_Improv into master

Hi everybody,

this MR introduces a couple of selection additional selection cuts on the quadruplets entering the FourLepton vertex fit. The fit is only performed if:

  • The di-lepton pair with the highest mass has to be at least 40 GeV regardless of the flavour and charge composition
  • ee and mumu pairs with opposite charges have to be at least 3.5 GeV
  • The two leptons leading in pt need to exceed 15 and 10 GeV, respectively.
  • The four leptons have to be separated in z by less than 7.5 mm
  • Apply minimal quality criteria on the ID tracks as well.

Additionally the fitted vertex has to have a reduced chi2 of less than 20. The covariances and trackWeights are not needed in the analysis and hence their information is pruned from the vertex.

The implemented selections reduce the average number of fits per event from 25 to 0.5 and the disk resource usage from 60% to less than 1%. This will make our friend here very happy:

image

Tagging: @goblirsc, @sabidi, @carcange, @gfrattar

Edited by Johannes Junggeburth

Merge request reports