Home

• Analysis strategy
  The main purpose of the framework is to study the performances of the MET filters.
  Generally, the performances of the filters can be divided into two different kinds of performances such as

  • How much a filter is good at not rejecting physical events indicating its efficiency.
  • How much a filter is good at rejecting unphysical events (Noise rejection fraction).

  To have a desirable estimation of these performances, we need to study each performance in proper phase space. Therefore, we need to specify these control regions:

  1- Noise Free Region (NFR)

  2- Background Enriched Region (BER)

  We also need to define the performances according to these control regions.

  In NFR, as aforementioned, the performance of a filter is it's efficiency indeed, so we define the efficiency:

  

  In BER, the noise jection fraction is defined as,

  

  In which the Numerator and the Denominator are histograms filled by events passing the filter and filled by all the events respectively.

• N-1 Method
  There is an option (by default is ON) to put another global constraint on filling the Numerators and Denominators.
  This option is provided to apply the N-1 method in studying the performances of the filters.
  What is the N-1 method?
  By this method, we mean, for a filter understudy, the Numerators and Denominators are filled only if the event passes all the recommended MET filters except the filter understudy and all the updated versions of the filter understudy.

• Parameters
  In both control regions, the performances of filters are studied according to 4 different variables:
  • PF MET
  • PUPPI MET
  • Leading Jet Pt
  • Number Primary Vertices