file analyses/Analysis_CMS_13TeV_0LEP_137invfb.cpp
[No description available]
Namespaces
| Name |
|---|
| Gambit TODO: see if we can use this one: |
| Gambit::ColliderBit |
Classes
| Name | |
|---|---|
| class | Gambit::ColliderBit::Analysis_CMS_13TeV_0LEP_137invfb CMS Run 2 0-lepton jet+MET SUSY analysis, with 137/fb of data. |
Source code
// -*- C++ -*-
#include "gambit/ColliderBit/analyses/Analysis.hpp"
#include "gambit/ColliderBit/analyses/Cutflow.hpp"
#include "gambit/ColliderBit/CMSEfficiencies.hpp"
// #define CHECK_CUTFLOW
namespace Gambit
{
namespace ColliderBit
{
using namespace std;
using namespace HEPUtils;
/// @brief CMS Run 2 0-lepton jet+MET SUSY analysis, with 137/fb of data
///
/// Based on: http://cms-results.web.cern.ch/cms-results/public-results/publications/SUS-19-006/index.html
///
class Analysis_CMS_13TeV_0LEP_137invfb : public Analysis
{
public:
// Required detector sim
static constexpr const char* detector = "CMS";
// Numbers passing cuts
std::map<string, EventCounter> _counters = {
{"SR1", EventCounter("SR1")},
{"SR2", EventCounter("SR2")},
{"SR3", EventCounter("SR3")},
{"SR4", EventCounter("SR4")},
{"SR5", EventCounter("SR5")},
{"SR6", EventCounter("SR6")},
{"SR7", EventCounter("SR7")},
{"SR8", EventCounter("SR8")},
{"SR9", EventCounter("SR9")},
{"SR10", EventCounter("SR10")},
{"SR11", EventCounter("SR11")},
{"SR12", EventCounter("SR12")},
};
static const size_t NUMSR = 12;
Cutflow _cutflow;
Analysis_CMS_13TeV_0LEP_137invfb() :
_cutflow("CMS 0-lep 13 TeV",
{"Njet >= 2", "HT > 300", "HTmiss > 300", "HTmiss/HT < 1",
"Nmuon = 0", "Nelectron = 0", "Nphoton = 0",
"Dphi_htmiss_j1", "Dphi_htmiss_j2", "Dphi_htmiss_j3", "Dphi_htmiss_j4",
"Evt quality"}) //, "SR HTmiss", "SR HT", "SR Njet", "SR Nbjet"}
{
set_analysis_name("CMS_13TeV_0LEP_137invfb");
set_luminosity(137.0);
}
void run(const Event* event)
{
_cutflow.fillinit();
// Get jets
vector<const Jet*> jets24, jets50;
for (const Jet* jet : event->jets("antikt_R04"))
{
if (jet->pT() < 30) continue;
if (jet->abseta() < 2.4) jets24.push_back(jet);
if (jet->abseta() < 5.0) jets50.push_back(jet);
}
const size_t njets = jets24.size();
if (njets < 2) return;
_cutflow.fill(1);
// Count b-jets
size_t nbjets = 0;
for (const Jet* j : jets24)
if (random_bool(j->btag() ? 0.65 : j->ctag() ? 0.13 : 0.016))
nbjets += 1;
// HT cut
double sumptj = 0;
for (const Jet* j : jets24) sumptj += j->pT();
const double ht = sumptj;
if (ht < 300) return;
_cutflow.fill(2);
// HTmiss cut, from full set of jets
P4 htvec;
for (const Jet* jet : jets50) htvec += jet->mom();
const double htmiss = htvec.pT();
if (htmiss < 300) return;
_cutflow.fill(3);
// HTmiss/HT cut
if (htmiss/ht >= 1) return;
_cutflow.fill(4);
// Get baseline photons
vector<const Particle*> basephotons;
for (const Particle* gamma : event->photons())
if (gamma->pT() > 10. && gamma->abseta() < 2.4)
basephotons.push_back(gamma);
// Get baseline electrons and apply efficiency
vector<const Particle*> baseelecs;
for (const Particle* electron : event->electrons())
if (electron->pT() > 10. && electron->abseta() < 2.5)
baseelecs.push_back(electron);
CMS::applyElectronEff(baseelecs);
// Get baseline muons and apply efficiency
vector<const Particle*> basemuons;
for (const Particle* muon : event->muons())
if (muon->pT() > 10. && muon->abseta() < 2.4)
basemuons.push_back(muon);
CMS::applyMuonEff(basemuons);
// Photon isolation
/// @todo Sum should actually be over all calo particles
vector<const Particle*> photons;
for (const Particle* y : basephotons)
{
const double R = 0.3;
double sumpt = -y->pT();
for (const Jet* j : jets50)
if (y->mom().deltaR_eta(j->mom()) < R) sumpt += j->pT();
if (sumpt/y->pT() < 0.1) photons.push_back(y); //< guess at threshold: real one not given, and varies between endcap/barrel
}
// Electron isolation
/// @todo Sum should actually be over all non-e/mu calo particles
vector<const Particle*> elecs;
for (const Particle* e : baseelecs)
{
const double R = max(0.05, min(0.2, 10/e->pT()));
double sumpt = -e->pT();
for (const Jet* j : jets50)
if (e->mom().deltaR_eta(j->mom()) < R) sumpt += j->pT();
if (sumpt/e->pT() < 0.1) elecs.push_back(e);
}
// Muon isolation
/// @todo Sum should actually be over all non-e/mu calo particles
vector<const Particle*> muons;
for (const Particle* m : basemuons)
{
const double R = max(0.05, min(0.2, 10/m->pT()));
double sumpt = -m->pT();
for (const Jet* j : jets50)
if (m->mom().deltaR_eta(j->mom()) < R) sumpt += j->pT();
if (sumpt/m->pT() < 0.2) muons.push_back(m);
}
// Veto the event if there are any remaining baseline leptons
if (!muons.empty()) return;
_cutflow.fill(5);
if (!elecs.empty()) return;
_cutflow.fill(6);
// Veto high-pT photons (should have negligible effect)
if (!photons.empty() && photons[0]->pT() > 100) return;
_cutflow.fill(7);
// Lead jets isolation from Htmiss
if (deltaPhi(-htvec, jets24[0]->mom()) < 0.5) return;
_cutflow.fill(8);
if (deltaPhi(-htvec, jets24[1]->mom()) < 0.5) return;
_cutflow.fill(9);
if (jets24.size() >= 3 && deltaPhi(-htvec, jets24[2]->mom()) < 0.3) return;
_cutflow.fill(10);
if (jets24.size() >= 4 && deltaPhi(-htvec, jets24[3]->mom()) < 0.3) return;
_cutflow.fill(11);
// Downweight for event quality inefficiency
const double w = 0.95 * event->weight();
const double werr = 0.95 * event->weight_err();
if (random_bool(0.95)) _cutflow.fill(12);
// Fill aggregate SR bins
if (htmiss >= 600 && ht >= 600 && njets >= 2 && nbjets == 0) _counters.at("SR1").add_event(w,werr);
if (htmiss >= 850 && ht >= 1700 && njets >= 4 && nbjets == 0) _counters.at("SR2").add_event(w,werr);
if (htmiss >= 600 && ht >= 600 && njets >= 6 && nbjets == 0) _counters.at("SR3").add_event(w,werr);
if (htmiss >= 600 && ht >= 600 && njets >= 8 && nbjets <= 1) _counters.at("SR4").add_event(w,werr);
if (htmiss >= 850 && ht >= 1700 && njets >= 10 && nbjets <= 1) _counters.at("SR5").add_event(w,werr);
if (htmiss >= 300 && ht >= 300 && njets >= 4 && nbjets >= 2) _counters.at("SR6").add_event(w,werr);
if (htmiss >= 600 && ht >= 600 && njets >= 2 && nbjets >= 2) _counters.at("SR7").add_event(w,werr);
if (htmiss >= 350 && ht >= 350 && njets >= 6 && nbjets >= 2) _counters.at("SR8").add_event(w,werr);
if (htmiss >= 600 && ht >= 600 && njets >= 4 && nbjets >= 2) _counters.at("SR9").add_event(w,werr);
if (htmiss >= 300 && ht >= 300 && njets >= 8 && nbjets >= 3) _counters.at("SR10").add_event(w,werr);
if (htmiss >= 600 && ht >= 600 && njets >= 6 && nbjets >= 1) _counters.at("SR11").add_event(w,werr);
if (htmiss >= 850 && ht >= 850 && njets >= 10 && nbjets >= 3) _counters.at("SR12").add_event(w,werr);
}
/// Combine the variables of another copy of this analysis (typically on another thread) into this one.
void combine(const Analysis* other)
{
const Analysis_CMS_13TeV_0LEP_137invfb* specificOther = dynamic_cast<const Analysis_CMS_13TeV_0LEP_137invfb*>(other);
for (auto& pair : _counters) { pair.second += specificOther->_counters.at(pair.first); }
}
/// Register results objects with the results for each SR; obs & bkg numbers from the CONF note
void collect_results()
{
add_result(SignalRegionData(_counters.at("SR1"), 11281., {12319., add_quad(85., 450.)} ));
add_result(SignalRegionData(_counters.at("SR2"), 74., {65.8, add_quad(6.0, 4.9)} ));
add_result(SignalRegionData(_counters.at("SR3"), 505., {489., add_quad(15., 18.)} ));
add_result(SignalRegionData(_counters.at("SR4"), 63., {54.3, add_quad(5.5, 2.5)} ));
add_result(SignalRegionData(_counters.at("SR5"), 153., {141., add_quad(10., 9.)} ));
add_result(SignalRegionData(_counters.at("SR6"), 10216., {10091., add_quad(115., 330.)} ));
add_result(SignalRegionData(_counters.at("SR7"), 287., {336., add_quad(15., 26.)} ));
add_result(SignalRegionData(_counters.at("SR8"), 1637., {1720., add_quad(35., 47.)} ));
add_result(SignalRegionData(_counters.at("SR9"), 224., {230., add_quad(13., 12.)} ));
add_result(SignalRegionData(_counters.at("SR10"), 168., {176., add_quad(11., 11.)} ));
add_result(SignalRegionData(_counters.at("SR11"), 282., {304., add_quad(14., 16.)} ));
add_result(SignalRegionData(_counters.at("SR12"), 0., {0.1, add_quad(1.2, 0.1)} ));
// Cutflow printout
#ifdef CHECK_CUTFLOW
// const double sf = 137*crossSection()/femtobarn/sumOfWeights();
// _cutflows.scale(sf);
cout << "\nCUTFLOWS:\n" << _cutflow << "\n" << endl;
cout << "\nSRCOUNTS:\n";
// Note: The sum() call below gives the raw event count. Use weight_sum() for the sum of event weights.
for (auto& pair : _counters) cout << pair.second.sum() << " ";
cout << "\n" << endl;
#endif
}
protected:
void analysis_specific_reset()
{
for (auto& pair : _counters) { pair.second.reset(); }
}
};
// Factory fn
DEFINE_ANALYSIS_FACTORY(CMS_13TeV_0LEP_137invfb)
}
}
Updated on 2024-07-18 at 13:53:34 +0000