Trigger studies for DM signals

IncandelaLab / LDMX-scripts

4 stars 15 forks source link

Trigger studies for DM signals #29

Open tvami opened 6 months ago

tvami commented 6 months ago

Issue description

Tasks

[ ] Look at pT vs eff for signal
[ ] Look at MET vs eff for signal

Existing material to start off

One handy way to do this is to fill a 2D histogram in ROOT with one axis having 2 bins (-0.5-0.5 and 0.5-1.5) filling 0 when the trigger is not fired and filling 1 when it is; then the other axis is the parameter of interest. Then take a TProfile in the boolean axis and that's it

DuncanWilmot commented 6 months ago

FYI it turns out trigger efficiencies are more like this (monatomic increase could be explained by increasing missing energy even if no pT bias):

Trigger efficiencies:{
    "0.001": 0.7723346153846153,
    "0.01": 0.8991299107142857,
    "0.1": 0.9228279816513761,
    "1.0": 0.9370799065420561,
}

I can make pT and ME plots before Wednesday if this is time-sensitive (also LDMX can measure true total missing energy, but maybe MET calculations still possible/useful?)

tvami commented 6 months ago

Thanks @DuncanWilmot , this makes more sense indeed! It's not time sensitive and also could be a small project from one of the incoming undergrads.