L1ECAL prefire issue

[ram1123]

Issue

1. Signal yield changes by ~10% after implementing the L1ECAL prefire weight.

---

Implemented here: https://github.com/ram1123/WWAnalysisRun2/blob/e5411f6ca4b299892fbba86a0826c03080ccee92/bin/produceWWNtuples.cpp#L1301-L1401

---

Yield before and after prefire weight

Total % change in MC ~5%, whiel for WV_EWK % change is ~10%.

Yield before L1ECAL prefire weight

Total MC	= 333.432.            
data	= 347
data/MC	= 1.04069
WV_EWK	= 17.8229
Diboson	= 33.6205
W+jets	= 180.266
Z+jets	= 1.42361
top	= 118.122

Yield after L1ECAL prefire weight

Total MC	= 314.123
data	= 347
data/MC	= 1.10466
WV_EWK	= 15.8743
Diboson	= 31.5216
W+jets	= 169.683
Z+jets	= 1.42361
top	= 111.494

[ram1123]

Jets contribution to ECAL prefire

WV_EWK	=  16.196
Diboson	= 31.9591
W+jets	= 171.525
Z+jets	= 1.42361
top	= 112.865

Now signal changes by ~9%

---

code patch

https://github.com/ram1123/WWAnalysisRun2/blob/82b9a3e462febaba7fe39228d56883825af12371/bin/produceWWNtuples.cpp#L1306-L1340

[ram1123]

Photons contribution only to WV EWK

WV_EWK	= 17.2104

signal yield changes by 3%.

Yield after prefire weight (Clean photons & jets not associated with photons)

Total MC	= 319.992
data	= 347
data/MC	= 1.0844
WV_EWK	=  16.419
Diboson	=  32.191
W+jets	= 172.712
Z+jets	= 1.42361
top	= 113.665

% change in yield ~9%

Further Check

• See the data/mc before and after prefire weight of following variables.
  • mjj
  • VBFj1_eta
  • VBFj2_eta

[ram1123]

VBF H→inv reference

Slide Link: link

Slide-8: Says effect on VBF is at-least ~9%.

Data/MC comparison w/o L1ECAL prefire weight

With L1ECAL prefire (Jets && photons including photons associated with jets)

To increase the statistiscs mjj and dEtajj cuts are reduced to 400 and 2.5, respectively. Below data/mc shows that between 2< |eta| < 3 data/mc imporves.

No Prefire weight	Prefire weight