vector<Double_t*> simFit(bool makeSoupFit_ = false,
const string tnp_ = "etoTauMargLooseNoCracks70",
const string category_ = "tauAntiEMVA",
const string bin_ = "abseta<1.5",
const float binCenter_ = 0.75,
const float binWidth_ = 0.75,
const float xLow_=60,
const float xHigh_=120,
bool SumW2_ = false,
bool verbose_ = true){
vector<Double_t*> out;
//return out;
//TFile *test = new TFile( outFile->GetName(),"UPDATE");
// output file
TFile *test = new TFile( Form("EtoTauPlotsFit_%s_%s_%f.root",tnp_.c_str(),category_.c_str(),binCenter_),"RECREATE");
test->mkdir(Form("bin%f",binCenter_));
TCanvas *c = new TCanvas("fitCanvas",Form("fitCanvas_%s_%s",tnp_.c_str(),bin_.c_str()),10,30,650,600);
c->SetGrid(0,0);
c->SetFillStyle(4000);
c->SetFillColor(10);
c->SetTicky();
c->SetObjectStat(0);
TCanvas *c2 = new TCanvas("fitCanvasTemplate",Form("fitCanvasTemplate_%s_%s",tnp_.c_str(),bin_.c_str()),10,30,650,600);
c2->SetGrid(0,0);
c2->SetFillStyle(4000);
c2->SetFillColor(10);
c2->SetTicky();
c2->SetObjectStat(0);
// input files
TFile fsup("/data_CMS/cms/lbianchini/tagAndProbe/trees/38XWcut/testNewWriteFromPAT_soup.root");
TFile fbkg("/data_CMS/cms/lbianchini/tagAndProbe/trees/38XWcut/testNewWriteFromPAT_soup_bkg.root");
TFile fsgn("/data_CMS/cms/lbianchini/tagAndProbe/trees/38XWcut/testNewWriteFromPAT_soup_sgn.root");
TFile fdat("/data_CMS/cms/lbianchini/tagAndProbe/trees/38XWcut/testNewWriteFromPAT_Data.root");
// data from 2iter:
//TFile fdat("/data_CMS/cms/lbianchini/35pb/testNewWriteFromPAT_Data.root");
//********************** signal only tree *************************/
TTree *fullTreeSgn = (TTree*)fsgn.Get((tnp_+"/fitter_tree").c_str());
TH1F* hSall = new TH1F("hSall","",1,0,150);
TH1F* hSPall = new TH1F("hSPall","",1,0,150);
TH1F* hS = new TH1F("hS","",1,0,150);
TH1F* hSP = new TH1F("hSP","",1,0,150);
fullTreeSgn->Draw("mass>>hS",Form("weight*(%s && mass>%f && mass<%f && mcTrue && signalPFChargedHadrCands<1.5)",bin_.c_str(),xLow_,xHigh_));
fullTreeSgn->Draw("mass>>hSall",Form("weight*(%s && mass>%f && mass<%f)",bin_.c_str(),xLow_,xHigh_));
float SGNtrue = hS->Integral();
float SGNall = hSall->Integral();
fullTreeSgn->Draw("mass>>hSP",Form("weight*(%s && %s>0 && mass>%f && mass<%f && mcTrue && signalPFChargedHadrCands<1.5 )",bin_.c_str(),category_.c_str(),xLow_,xHigh_));
fullTreeSgn->Draw("mass>>hSPall",Form("weight*(%s && %s>0 && mass>%f && mass<%f && signalPFChargedHadrCands<1.5 )",bin_.c_str(),category_.c_str(),xLow_,xHigh_));
float SGNtruePass = hSP->Integral();
float SGNallPass = hSPall->Integral();
//********************** background only tree *************************//
TTree *fullTreeBkg = (TTree*)fbkg.Get((tnp_+"/fitter_tree").c_str());
TH1F* hB = new TH1F("hB","",1,0,150);
TH1F* hBP = new TH1F("hBP","",1,0,150);
fullTreeBkg->Draw("mass>>hB",Form("weight*(%s && mass>%f && mass<%f && signalPFChargedHadrCands<1.5 )",bin_.c_str(),xLow_,xHigh_));
float BKG = hB->Integral();
float BKGUnWeighted = hB->GetEntries();
fullTreeBkg->Draw("mass>>hBP",Form("weight*(%s && %s>0 && mass>%f && mass<%f && signalPFChargedHadrCands<1.5 )",bin_.c_str(),category_.c_str(),xLow_,xHigh_));
float BKGPass = hBP->Integral();
float BKGUnWeightedPass = hBP->GetEntries();
float BKGFail = BKG-BKGPass;
cout << "*********** BKGFail " << BKGFail << endl;
//********************** soup tree *************************//
TTree *fullTreeSoup = (TTree*)fsup.Get((tnp_+"/fitter_tree").c_str());
//********************** data tree *************************//
TTree *fullTreeData = (TTree*)fdat.Get((tnp_+"/fitter_tree").c_str());
//********************** workspace ***********************//
RooWorkspace *w = new RooWorkspace("w","w");
// tree variables to be imported
w->factory("mass[30,120]");
w->factory("weight[0,10000]");
w->factory("abseta[0,2.5]");
w->factory("pt[0,200]");
w->factory("mcTrue[0,1]");
w->factory("signalPFChargedHadrCands[0,10]");
w->factory((category_+"[0,1]").c_str());
// background pass pdf for MC
w->factory("RooExponential::McBackgroundPdfP(mass,McCP[0,-10,10])");
// background fail pdf for MC
w->factory("RooExponential::McBackgroundPdfF(mass,McCF[0,-10,10])");
// background pass pdf for Data
w->factory("RooExponential::DataBackgroundPdfP(mass,DataCP[0,-10,10])");
// background fail pdf for Data
w->factory("RooExponential::DataBackgroundPdfF(mass,DataCF[0,-10,10])");
// fit parameters for background
w->factory("McEfficiency[0.04,0,1]");
w->factory("McNumSgn[0,1000000]");
w->factory("McNumBkgP[0,100000]");
w->factory("McNumBkgF[0,100000]");
w->factory("expr::McNumSgnP('McEfficiency*McNumSgn',McEfficiency,McNumSgn)");
w->factory("expr::McNumSgnF('(1-McEfficiency)*McNumSgn',McEfficiency,McNumSgn)");
w->factory("McPassing[pass=1,fail=0]");
// fit parameters for data
w->factory("DataEfficiency[0.1,0,1]");
w->factory("DataNumSgn[0,1000000]");
w->factory("DataNumBkgP[0,1000000]");
w->factory("DataNumBkgF[0,10000]");
w->factory("expr::DataNumSgnP('DataEfficiency*DataNumSgn',DataEfficiency,DataNumSgn)");
w->factory("expr::DataNumSgnF('(1-DataEfficiency)*DataNumSgn',DataEfficiency,DataNumSgn)");
w->factory("DataPassing[pass=1,fail=0]");
RooRealVar *weight = w->var("weight");
RooRealVar *abseta = w->var("abseta");
RooRealVar *pt = w->var("pt");
RooRealVar *mass = w->var("mass");
mass->setRange(xLow_,xHigh_);
RooRealVar *mcTrue = w->var("mcTrue");
RooRealVar *cut = w->var( category_.c_str() );
RooRealVar *signalPFChargedHadrCands = w->var("signalPFChargedHadrCands");
// build the template for the signal pass sample:
RooDataSet templateP("templateP","dataset for signal-pass template", RooArgSet(*mass,*weight,*abseta,*pt,*cut,*mcTrue,*signalPFChargedHadrCands), Import( *fullTreeSgn ), /*WeightVar( *weight ),*/ Cut( Form("(mcTrue && %s>0.5 && %s && signalPFChargedHadrCands<1.5)",category_.c_str(),bin_.c_str()) ) );
// build the template for the signal fail sample:
RooDataSet templateF("templateF","dataset for signal-fail template", RooArgSet(*mass,*weight,*abseta,*pt,*cut,*mcTrue,*signalPFChargedHadrCands), Import( *fullTreeSgn ), /*WeightVar( *weight ),*/ Cut( Form("(mcTrue && %s<0.5 && %s && signalPFChargedHadrCands<1.5)",category_.c_str(),bin_.c_str()) ) );
mass->setBins(24);
RooDataHist templateHistP("templateHistP","",RooArgSet(*mass), templateP, 1.0);
RooHistPdf TemplateSignalPdfP("TemplateSignalPdfP","",RooArgSet(*mass),templateHistP);
w->import(TemplateSignalPdfP);
mass->setBins(24);
RooDataHist templateHistF("templateHistF","",RooArgSet(*mass),templateF,1.0);
RooHistPdf TemplateSignalPdfF("TemplateSignalPdfF","",RooArgSet(*mass),templateHistF);
w->import(TemplateSignalPdfF);
mass->setBins(10000,"fft");
RooPlot* TemplateFrameP = mass->frame(Bins(24),Title("Template passing"));
templateP.plotOn(TemplateFrameP);
w->pdf("TemplateSignalPdfP")->plotOn(TemplateFrameP);
RooPlot* TemplateFrameF = mass->frame(Bins(24),Title("Template failing"));
templateF.plotOn(TemplateFrameF);
w->pdf("TemplateSignalPdfF")->plotOn(TemplateFrameF);
//w->factory("RooFFTConvPdf::McSignalPdfP(mass,TemplateSignalPdfP,RooTruthModel::McResolModP(mass))");
//w->factory("RooFFTConvPdf::McSignalPdfF(mass,TemplateSignalPdfF,RooTruthModel::McResolModF(mass))");
// FOR GREGORY: PROBLEM WHEN TRY TO USE THE PURE TEMPLATE =>
RooHistPdf McSignalPdfP("McSignalPdfP","McSignalPdfP",RooArgSet(*mass),templateHistP);
RooHistPdf McSignalPdfF("McSignalPdfF","McSignalPdfF",RooArgSet(*mass),templateHistF);
w->import(McSignalPdfP);
w->import(McSignalPdfF);
// FOR GREGORY: FOR DATA, CONVOLUTION IS OK =>
w->factory("RooFFTConvPdf::DataSignalPdfP(mass,TemplateSignalPdfP,RooGaussian::DataResolModP(mass,DataMeanResP[0.0,-5.,5.],DataSigmaResP[0.5,0.,10]))");
w->factory("RooFFTConvPdf::DataSignalPdfF(mass,TemplateSignalPdfF,RooGaussian::DataResolModF(mass,DataMeanResF[-5.,-10.,10.],DataSigmaResF[0.5,0.,10]))");
//w->factory("RooCBShape::DataSignalPdfF(mass,DataMeanF[91.2,88,95.],DataSigmaF[3,0.5,8],DataAlfaF[1.8,0.,10],DataNF[1.0,1e-06,10])");
//w->factory("RooFFTConvPdf::DataSignalPdfF(mass,RooVoigtian::DataVoigF(mass,DataMeanF[85,80,95],DataWidthF[2.49],DataSigmaF[3,0.5,10]),RooCBShape::DataResolModF(mass,DataMeanResF[0.5,0.,10.],DataSigmaResF[0.5,0.,10],DataAlphaResF[0.5,0.,10],DataNResF[1.0,1e-06,10]))");
//w->factory("SUM::DataSignalPdfF(fVBP[0.5,0,1]*RooBifurGauss::bifF(mass,DataMeanResF[91.2,80,95],sigmaLF[10,0.5,40],sigmaRF[0.]), RooVoigtian::voigF(mass, DataMeanResF, widthF[2.49], sigmaVoigF[5,0.1,10]) )" );
// composite model pass for MC
w->factory("SUM::McModelP(McNumSgnP*McSignalPdfP,McNumBkgP*McBackgroundPdfP)");
w->factory("SUM::McModelF(McNumSgnF*McSignalPdfF,McNumBkgF*McBackgroundPdfF)");
// composite model pass for data
w->factory("SUM::DataModelP(DataNumSgnP*DataSignalPdfP,DataNumBkgP*DataBackgroundPdfP)");
w->factory("SUM::DataModelF(DataNumSgnF*DataSignalPdfF,DataNumBkgF*DataBackgroundPdfF)");
// simultaneous fir for MC
w->factory("SIMUL::McModel(McPassing,pass=McModelP,fail=McModelF)");
// simultaneous fir for data
w->factory("SIMUL::DataModel(DataPassing,pass=DataModelP,fail=DataModelF)");
w->Print("V");
w->saveSnapshot("clean", w->allVars());
w->loadSnapshot("clean");
/****************** sim fit to soup **************************/
///////////////////////////////////////////////////////////////
TFile *f = new TFile("dummySoup.root","RECREATE");
TTree* cutTreeSoupP = fullTreeSoup->CopyTree(Form("(%s>0.5 && %s && signalPFChargedHadrCands<1.5)",category_.c_str(),bin_.c_str()));
TTree* cutTreeSoupF = fullTreeSoup->CopyTree(Form("(%s<0.5 && %s && signalPFChargedHadrCands<1.5)",category_.c_str(),bin_.c_str()));
RooDataSet McDataP("McDataP","dataset pass for the soup", RooArgSet(*mass), Import( *cutTreeSoupP ) );
RooDataSet McDataF("McDataF","dataset fail for the soup", RooArgSet(*mass), Import( *cutTreeSoupF ) );
RooDataHist McCombData("McCombData","combined data for the soup", RooArgSet(*mass), Index(*(w->cat("McPassing"))), Import("pass", *(McDataP.createHistogram("histoP",*mass)) ), Import("fail",*(McDataF.createHistogram("histoF",*mass)) ) ) ;
RooPlot* McFrameP = 0;
RooPlot* McFrameF = 0;
RooRealVar* McEffFit = 0;
if(makeSoupFit_){
cout << "**************** N bins in mass " << w->var("mass")->getBins() << endl;
RooFitResult* ResMcCombinedFit = w->pdf("McModel")->fitTo(McCombData, Extended(1), Minos(1), Save(1), SumW2Error( SumW2_ ), Range(xLow_,xHigh_), NumCPU(4) /*, ExternalConstraints( *(w->pdf("ConstrainMcNumBkgF")) )*/ );
test->cd(Form("bin%f",binCenter_));
ResMcCombinedFit->Write("McFitResults_Combined");
RooArgSet McFitParam(ResMcCombinedFit->floatParsFinal());
McEffFit = (RooRealVar*)(&McFitParam["McEfficiency"]);
RooRealVar* McNumSigFit = (RooRealVar*)(&McFitParam["McNumSgn"]);
RooRealVar* McNumBkgPFit = (RooRealVar*)(&McFitParam["McNumBkgP"]);
RooRealVar* McNumBkgFFit = (RooRealVar*)(&McFitParam["McNumBkgF"]);
McFrameP = mass->frame(Bins(24),Title("MC: passing sample"));
McCombData.plotOn(McFrameP,Cut("McPassing==McPassing::pass"));
w->pdf("McModel")->plotOn(McFrameP,Slice(*(w->cat("McPassing")),"pass"), ProjWData(*(w->cat("McPassing")),McCombData), LineColor(kBlue),Range(xLow_,xHigh_));
w->pdf("McModel")->plotOn(McFrameP,Slice(*(w->cat("McPassing")),"pass"), ProjWData(*(w->cat("McPassing")),McCombData), Components("McSignalPdfP"), LineColor(kRed),Range(xLow_,xHigh_));
w->pdf("McModel")->plotOn(McFrameP,Slice(*(w->cat("McPassing")),"pass"), ProjWData(*(w->cat("McPassing")),McCombData), Components("McBackgroundPdfP"), LineColor(kGreen),Range(xLow_,xHigh_));
McFrameF = mass->frame(Bins(24),Title("MC: failing sample"));
McCombData.plotOn(McFrameF,Cut("McPassing==McPassing::fail"));
w->pdf("McModel")->plotOn(McFrameF,Slice(*(w->cat("McPassing")),"fail"), ProjWData(*(w->cat("McPassing")),McCombData), LineColor(kBlue),Range(xLow_,xHigh_));
w->pdf("McModel")->plotOn(McFrameF,Slice(*(w->cat("McPassing")),"fail"), ProjWData(*(w->cat("McPassing")),McCombData), Components("McSignalPdfF"), LineColor(kRed),Range(xLow_,xHigh_));
w->pdf("McModel")->plotOn(McFrameF,Slice(*(w->cat("McPassing")),"fail"), ProjWData(*(w->cat("McPassing")),McCombData), Components("McBackgroundPdfF"), LineColor(kGreen),Range(xLow_,xHigh_));
}
///////////////////////////////////////////////////////////////
/****************** sim fit to data **************************/
///////////////////////////////////////////////////////////////
TFile *f2 = new TFile("dummyData.root","RECREATE");
TTree* cutTreeDataP = fullTreeData->CopyTree(Form("(%s>0.5 && %s && signalPFChargedHadrCands<1.5)",category_.c_str(),bin_.c_str()));
TTree* cutTreeDataF = fullTreeData->CopyTree(Form("(%s<0.5 && %s && signalPFChargedHadrCands<1.5)",category_.c_str(),bin_.c_str()));
RooDataSet DataDataP("DataDataP","dataset pass for the soup", RooArgSet(*mass), Import( *cutTreeDataP ) );
RooDataSet DataDataF("DataDataF","dataset fail for the soup", RooArgSet(*mass), Import( *cutTreeDataF ) );
RooDataHist DataCombData("DataCombData","combined data for the soup", RooArgSet(*mass), Index(*(w->cat("DataPassing"))), Import("pass",*(DataDataP.createHistogram("histoDataP",*mass))),Import("fail",*(DataDataF.createHistogram("histoDataF",*mass)))) ;
RooFitResult* ResDataCombinedFit = w->pdf("DataModel")->fitTo(DataCombData, Extended(1), Minos(1), Save(1), SumW2Error( SumW2_ ), Range(xLow_,xHigh_), NumCPU(4));
test->cd(Form("bin%f",binCenter_));
ResDataCombinedFit->Write("DataFitResults_Combined");
RooArgSet DataFitParam(ResDataCombinedFit->floatParsFinal());
RooRealVar* DataEffFit = (RooRealVar*)(&DataFitParam["DataEfficiency"]);
RooRealVar* DataNumSigFit = (RooRealVar*)(&DataFitParam["DataNumSgn"]);
RooRealVar* DataNumBkgPFit = (RooRealVar*)(&DataFitParam["DataNumBkgP"]);
RooRealVar* DataNumBkgFFit = (RooRealVar*)(&DataFitParam["DataNumBkgF"]);
RooPlot* DataFrameP = mass->frame(Bins(24),Title("Data: passing sample"));
DataCombData.plotOn(DataFrameP,Cut("DataPassing==DataPassing::pass"));
w->pdf("DataModel")->plotOn(DataFrameP,Slice(*(w->cat("DataPassing")),"pass"), ProjWData(*(w->cat("DataPassing")),DataCombData), LineColor(kBlue),Range(xLow_,xHigh_));
w->pdf("DataModel")->plotOn(DataFrameP,Slice(*(w->cat("DataPassing")),"pass"), ProjWData(*(w->cat("DataPassing")),DataCombData), Components("DataSignalPdfP"), LineColor(kRed),Range(xLow_,xHigh_));
w->pdf("DataModel")->plotOn(DataFrameP,Slice(*(w->cat("DataPassing")),"pass"), ProjWData(*(w->cat("DataPassing")),DataCombData), Components("DataBackgroundPdfP"), LineColor(kGreen),LineStyle(kDashed),Range(xLow_,xHigh_));
RooPlot* DataFrameF = mass->frame(Bins(24),Title("Data: failing sample"));
DataCombData.plotOn(DataFrameF,Cut("DataPassing==DataPassing::fail"));
w->pdf("DataModel")->plotOn(DataFrameF,Slice(*(w->cat("DataPassing")),"fail"), ProjWData(*(w->cat("DataPassing")),DataCombData), LineColor(kBlue),Range(xLow_,xHigh_));
w->pdf("DataModel")->plotOn(DataFrameF,Slice(*(w->cat("DataPassing")),"fail"), ProjWData(*(w->cat("DataPassing")),DataCombData), Components("DataSignalPdfF"), LineColor(kRed),Range(xLow_,xHigh_));
w->pdf("DataModel")->plotOn(DataFrameF,Slice(*(w->cat("DataPassing")),"fail"), ProjWData(*(w->cat("DataPassing")),DataCombData), Components("DataBackgroundPdfF"), LineColor(kGreen),LineStyle(kDashed),Range(xLow_,xHigh_));
///////////////////////////////////////////////////////////////
if(makeSoupFit_) c->Divide(2,2);
else c->Divide(2,1);
c->cd(1);
DataFrameP->Draw();
c->cd(2);
DataFrameF->Draw();
if(makeSoupFit_){
c->cd(3);
McFrameP->Draw();
c->cd(4);
McFrameF->Draw();
}
c->Draw();
test->cd(Form("bin%f",binCenter_));
c->Write();
c2->Divide(2,1);
c2->cd(1);
TemplateFrameP->Draw();
c2->cd(2);
TemplateFrameF->Draw();
c2->Draw();
test->cd(Form("bin%f",binCenter_));
c2->Write();
// MINOS errors, otherwise HESSE quadratic errors
float McErrorLo = 0;
float McErrorHi = 0;
if(makeSoupFit_){
McErrorLo = McEffFit->getErrorLo()<0 ? McEffFit->getErrorLo() : (-1)*McEffFit->getError();
McErrorHi = McEffFit->getErrorHi()>0 ? McEffFit->getErrorHi() : McEffFit->getError();
}
float DataErrorLo = DataEffFit->getErrorLo()<0 ? DataEffFit->getErrorLo() : (-1)*DataEffFit->getError();
float DataErrorHi = DataEffFit->getErrorHi()>0 ? DataEffFit->getErrorHi() : DataEffFit->getError();
float BinomialError = TMath::Sqrt(SGNtruePass/SGNtrue*(1-SGNtruePass/SGNtrue)/SGNtrue);
Double_t* truthMC = new Double_t[6];
Double_t* tnpMC = new Double_t[6];
Double_t* tnpData = new Double_t[6];
truthMC[0] = binCenter_;
truthMC[1] = binWidth_;
truthMC[2] = binWidth_;
truthMC[3] = SGNtruePass/SGNtrue;
truthMC[4] = BinomialError;
truthMC[5] = BinomialError;
if(makeSoupFit_){
tnpMC[0] = binCenter_;
tnpMC[1] = binWidth_;
tnpMC[2] = binWidth_;
tnpMC[3] = McEffFit->getVal();
tnpMC[4] = (-1)*McErrorLo;
tnpMC[5] = McErrorHi;
}
tnpData[0] = binCenter_;
tnpData[1] = binWidth_;
tnpData[2] = binWidth_;
tnpData[3] = DataEffFit->getVal();
tnpData[4] = (-1)*DataErrorLo;
tnpData[5] = DataErrorHi;
out.push_back(truthMC);
out.push_back(tnpData);
if(makeSoupFit_) out.push_back(tnpMC);
test->Close();
//delete c; delete c2;
if(verbose_) cout << "returning from bin " << bin_ << endl;
return out;
}
void OneSidedFrequentistUpperLimitWithBands(const char* infile = "",
const char* workspaceName = "combined",
const char* modelConfigName = "ModelConfig",
const char* dataName = "obsData") {
double confidenceLevel=0.95;
int nPointsToScan = 20;
int nToyMC = 200;
// -------------------------------------------------------
// First part is just to access a user-defined file
// or create the standard example file if it doesn't exist
const char* filename = "";
if (!strcmp(infile,"")) {
filename = "results/example_combined_GaussExample_model.root";
bool fileExist = !gSystem->AccessPathName(filename); // note opposite return code
// if file does not exists generate with histfactory
if (!fileExist) {
#ifdef _WIN32
cout << "HistFactory file cannot be generated on Windows - exit" << endl;
return;
#endif
// Normally this would be run on the command line
cout <<"will run standard hist2workspace example"<<endl;
gROOT->ProcessLine(".! prepareHistFactory .");
gROOT->ProcessLine(".! hist2workspace config/example.xml");
cout <<"\n\n---------------------"<<endl;
cout <<"Done creating example input"<<endl;
cout <<"---------------------\n\n"<<endl;
}
}
else
filename = infile;
// Try to open the file
TFile *file = TFile::Open(filename);
// if input file was specified byt not found, quit
if(!file ){
cout <<"StandardRooStatsDemoMacro: Input file " << filename << " is not found" << endl;
return;
}
// -------------------------------------------------------
// Now get the data and workspace
// get the workspace out of the file
RooWorkspace* w = (RooWorkspace*) file->Get(workspaceName);
if(!w){
cout <<"workspace not found" << endl;
return;
}
// get the modelConfig out of the file
ModelConfig* mc = (ModelConfig*) w->obj(modelConfigName);
// get the modelConfig out of the file
RooAbsData* data = w->data(dataName);
// make sure ingredients are found
if(!data || !mc){
w->Print();
cout << "data or ModelConfig was not found" <<endl;
return;
}
// -------------------------------------------------------
// Now get the POI for convenience
// you may want to adjust the range of your POI
RooRealVar* firstPOI = (RooRealVar*) mc->GetParametersOfInterest()->first();
/* firstPOI->setMin(0);*/
/* firstPOI->setMax(10);*/
// --------------------------------------------
// Create and use the FeldmanCousins tool
// to find and plot the 95% confidence interval
// on the parameter of interest as specified
// in the model config
// REMEMBER, we will change the test statistic
// so this is NOT a Feldman-Cousins interval
FeldmanCousins fc(*data,*mc);
fc.SetConfidenceLevel(confidenceLevel);
/* fc.AdditionalNToysFactor(0.25); // degrade/improve sampling that defines confidence belt*/
/* fc.UseAdaptiveSampling(true); // speed it up a bit, don't use for expected limits*/
fc.SetNBins(nPointsToScan); // set how many points per parameter of interest to scan
fc.CreateConfBelt(true); // save the information in the belt for plotting
// -------------------------------------------------------
// Feldman-Cousins is a unified limit by definition
// but the tool takes care of a few things for us like which values
// of the nuisance parameters should be used to generate toys.
// so let's just change the test statistic and realize this is
// no longer "Feldman-Cousins" but is a fully frequentist Neyman-Construction.
/* ProfileLikelihoodTestStatModified onesided(*mc->GetPdf());*/
/* fc.GetTestStatSampler()->SetTestStatistic(&onesided);*/
/* ((ToyMCSampler*) fc.GetTestStatSampler())->SetGenerateBinned(true); */
ToyMCSampler* toymcsampler = (ToyMCSampler*) fc.GetTestStatSampler();
ProfileLikelihoodTestStat* testStat = dynamic_cast<ProfileLikelihoodTestStat*>(toymcsampler->GetTestStatistic());
testStat->SetOneSided(true);
// Since this tool needs to throw toy MC the PDF needs to be
// extended or the tool needs to know how many entries in a dataset
// per pseudo experiment.
// In the 'number counting form' where the entries in the dataset
// are counts, and not values of discriminating variables, the
// datasets typically only have one entry and the PDF is not
// extended.
if(!mc->GetPdf()->canBeExtended()){
if(data->numEntries()==1)
fc.FluctuateNumDataEntries(false);
else
cout <<"Not sure what to do about this model" <<endl;
}
// We can use PROOF to speed things along in parallel
// However, the test statistic has to be installed on the workers
// so either turn off PROOF or include the modified test statistic
// in your `$ROOTSYS/roofit/roostats/inc` directory,
// add the additional line to the LinkDef.h file,
// and recompile root.
if (useProof) {
ProofConfig pc(*w, nworkers, "", false);
toymcsampler->SetProofConfig(&pc); // enable proof
}
if(mc->GetGlobalObservables()){
cout << "will use global observables for unconditional ensemble"<<endl;
mc->GetGlobalObservables()->Print();
toymcsampler->SetGlobalObservables(*mc->GetGlobalObservables());
}
// Now get the interval
PointSetInterval* interval = fc.GetInterval();
ConfidenceBelt* belt = fc.GetConfidenceBelt();
// print out the interval on the first Parameter of Interest
cout << "\n95% interval on " <<firstPOI->GetName()<<" is : ["<<
interval->LowerLimit(*firstPOI) << ", "<<
interval->UpperLimit(*firstPOI) <<"] "<<endl;
// get observed UL and value of test statistic evaluated there
RooArgSet tmpPOI(*firstPOI);
double observedUL = interval->UpperLimit(*firstPOI);
firstPOI->setVal(observedUL);
double obsTSatObsUL = fc.GetTestStatSampler()->EvaluateTestStatistic(*data,tmpPOI);
// Ask the calculator which points were scanned
RooDataSet* parameterScan = (RooDataSet*) fc.GetPointsToScan();
RooArgSet* tmpPoint;
// make a histogram of parameter vs. threshold
TH1F* histOfThresholds = new TH1F("histOfThresholds","",
parameterScan->numEntries(),
firstPOI->getMin(),
firstPOI->getMax());
histOfThresholds->GetXaxis()->SetTitle(firstPOI->GetName());
histOfThresholds->GetYaxis()->SetTitle("Threshold");
// loop through the points that were tested and ask confidence belt
// what the upper/lower thresholds were.
// For FeldmanCousins, the lower cut off is always 0
for(Int_t i=0; i<parameterScan->numEntries(); ++i){
tmpPoint = (RooArgSet*) parameterScan->get(i)->clone("temp");
//cout <<"get threshold"<<endl;
double arMax = belt->GetAcceptanceRegionMax(*tmpPoint);
double poiVal = tmpPoint->getRealValue(firstPOI->GetName()) ;
histOfThresholds->Fill(poiVal,arMax);
}
TCanvas* c1 = new TCanvas();
c1->Divide(2);
c1->cd(1);
histOfThresholds->SetMinimum(0);
histOfThresholds->Draw();
c1->cd(2);
// -------------------------------------------------------
// Now we generate the expected bands and power-constraint
// First: find parameter point for mu=0, with conditional MLEs for nuisance parameters
RooAbsReal* nll = mc->GetPdf()->createNLL(*data);
RooAbsReal* profile = nll->createProfile(*mc->GetParametersOfInterest());
firstPOI->setVal(0.);
profile->getVal(); // this will do fit and set nuisance parameters to profiled values
RooArgSet* poiAndNuisance = new RooArgSet();
if(mc->GetNuisanceParameters())
poiAndNuisance->add(*mc->GetNuisanceParameters());
poiAndNuisance->add(*mc->GetParametersOfInterest());
w->saveSnapshot("paramsToGenerateData",*poiAndNuisance);
RooArgSet* paramsToGenerateData = (RooArgSet*) poiAndNuisance->snapshot();
cout << "\nWill use these parameter points to generate pseudo data for bkg only" << endl;
paramsToGenerateData->Print("v");
RooArgSet unconditionalObs;
unconditionalObs.add(*mc->GetObservables());
unconditionalObs.add(*mc->GetGlobalObservables()); // comment this out for the original conditional ensemble
double CLb=0;
double CLbinclusive=0;
// Now we generate background only and find distribution of upper limits
TH1F* histOfUL = new TH1F("histOfUL","",100,0,firstPOI->getMax());
histOfUL->GetXaxis()->SetTitle("Upper Limit (background only)");
histOfUL->GetYaxis()->SetTitle("Entries");
for(int imc=0; imc<nToyMC; ++imc){
// set parameters back to values for generating pseudo data
// cout << "\n get current nuis, set vals, print again" << endl;
w->loadSnapshot("paramsToGenerateData");
// poiAndNuisance->Print("v");
RooDataSet* toyData = 0;
// now generate a toy dataset
if(!mc->GetPdf()->canBeExtended()){
if(data->numEntries()==1)
toyData = mc->GetPdf()->generate(*mc->GetObservables(),1);
else
cout <<"Not sure what to do about this model" <<endl;
} else{
// cout << "generating extended dataset"<<endl;
toyData = mc->GetPdf()->generate(*mc->GetObservables(),Extended());
}
// generate global observables
// need to be careful for simpdf
// RooDataSet* globalData = mc->GetPdf()->generate(*mc->GetGlobalObservables(),1);
RooSimultaneous* simPdf = dynamic_cast<RooSimultaneous*>(mc->GetPdf());
if(!simPdf){
RooDataSet *one = mc->GetPdf()->generate(*mc->GetGlobalObservables(), 1);
const RooArgSet *values = one->get();
RooArgSet *allVars = mc->GetPdf()->getVariables();
*allVars = *values;
delete allVars;
delete values;
delete one;
} else {
//try fix for sim pdf
TIterator* iter = simPdf->indexCat().typeIterator() ;
RooCatType* tt = NULL;
while((tt=(RooCatType*) iter->Next())) {
// Get pdf associated with state from simpdf
RooAbsPdf* pdftmp = simPdf->getPdf(tt->GetName()) ;
// Generate only global variables defined by the pdf associated with this state
RooArgSet* globtmp = pdftmp->getObservables(*mc->GetGlobalObservables()) ;
RooDataSet* tmp = pdftmp->generate(*globtmp,1) ;
// Transfer values to output placeholder
*globtmp = *tmp->get(0) ;
// Cleanup
delete globtmp ;
delete tmp ;
}
}
// globalData->Print("v");
// unconditionalObs = *globalData->get();
// mc->GetGlobalObservables()->Print("v");
// delete globalData;
// cout << "toy data = " << endl;
// toyData->get()->Print("v");
// get test stat at observed UL in observed data
firstPOI->setVal(observedUL);
double toyTSatObsUL = fc.GetTestStatSampler()->EvaluateTestStatistic(*toyData,tmpPOI);
// toyData->get()->Print("v");
// cout <<"obsTSatObsUL " <<obsTSatObsUL << "toyTS " << toyTSatObsUL << endl;
if(obsTSatObsUL < toyTSatObsUL) // not sure about <= part yet
CLb+= (1.)/nToyMC;
if(obsTSatObsUL <= toyTSatObsUL) // not sure about <= part yet
CLbinclusive+= (1.)/nToyMC;
// loop over points in belt to find upper limit for this toy data
double thisUL = 0;
for(Int_t i=0; i<parameterScan->numEntries(); ++i){
tmpPoint = (RooArgSet*) parameterScan->get(i)->clone("temp");
double arMax = belt->GetAcceptanceRegionMax(*tmpPoint);
firstPOI->setVal( tmpPoint->getRealValue(firstPOI->GetName()) );
// double thisTS = profile->getVal();
double thisTS = fc.GetTestStatSampler()->EvaluateTestStatistic(*toyData,tmpPOI);
// cout << "poi = " << firstPOI->getVal()
// << " max is " << arMax << " this profile = " << thisTS << endl;
// cout << "thisTS = " << thisTS<<endl;
if(thisTS<=arMax){
thisUL = firstPOI->getVal();
} else{
break;
}
}
/*
// loop over points in belt to find upper limit for this toy data
double thisUL = 0;
for(Int_t i=0; i<histOfThresholds->GetNbinsX(); ++i){
tmpPoint = (RooArgSet*) parameterScan->get(i)->clone("temp");
cout <<"---------------- "<<i<<endl;
tmpPoint->Print("v");
cout << "from hist " << histOfThresholds->GetBinCenter(i+1) <<endl;
double arMax = histOfThresholds->GetBinContent(i+1);
// cout << " threhold from Hist = aMax " << arMax<<endl;
// double arMax2 = belt->GetAcceptanceRegionMax(*tmpPoint);
// cout << "from scan arMax2 = "<< arMax2 << endl; // not the same due to TH1F not TH1D
// cout << "scan - hist" << arMax2-arMax << endl;
firstPOI->setVal( histOfThresholds->GetBinCenter(i+1));
// double thisTS = profile->getVal();
double thisTS = fc.GetTestStatSampler()->EvaluateTestStatistic(*toyData,tmpPOI);
// cout << "poi = " << firstPOI->getVal()
// << " max is " << arMax << " this profile = " << thisTS << endl;
// cout << "thisTS = " << thisTS<<endl;
// NOTE: need to add a small epsilon term for single precision vs. double precision
if(thisTS<=arMax + 1e-7){
thisUL = firstPOI->getVal();
} else{
break;
}
}
*/
histOfUL->Fill(thisUL);
// for few events, data is often the same, and UL is often the same
// cout << "thisUL = " << thisUL<<endl;
delete toyData;
}
histOfUL->Draw();
c1->SaveAs("one-sided_upper_limit_output.pdf");
// if you want to see a plot of the sampling distribution for a particular scan point:
/*
SamplingDistPlot sampPlot;
int indexInScan = 0;
tmpPoint = (RooArgSet*) parameterScan->get(indexInScan)->clone("temp");
firstPOI->setVal( tmpPoint->getRealValue(firstPOI->GetName()) );
toymcsampler->SetParametersForTestStat(tmpPOI);
SamplingDistribution* samp = toymcsampler->GetSamplingDistribution(*tmpPoint);
sampPlot.AddSamplingDistribution(samp);
sampPlot.Draw();
*/
// Now find bands and power constraint
Double_t* bins = histOfUL->GetIntegral();
TH1F* cumulative = (TH1F*) histOfUL->Clone("cumulative");
cumulative->SetContent(bins);
double band2sigDown, band1sigDown, bandMedian, band1sigUp,band2sigUp;
for(int i=1; i<=cumulative->GetNbinsX(); ++i){
if(bins[i]<RooStats::SignificanceToPValue(2))
band2sigDown=cumulative->GetBinCenter(i);
if(bins[i]<RooStats::SignificanceToPValue(1))
band1sigDown=cumulative->GetBinCenter(i);
if(bins[i]<0.5)
bandMedian=cumulative->GetBinCenter(i);
if(bins[i]<RooStats::SignificanceToPValue(-1))
band1sigUp=cumulative->GetBinCenter(i);
if(bins[i]<RooStats::SignificanceToPValue(-2))
band2sigUp=cumulative->GetBinCenter(i);
}
cout << "-2 sigma band " << band2sigDown << endl;
cout << "-1 sigma band " << band1sigDown << " [Power Constraint)]" << endl;
cout << "median of band " << bandMedian << endl;
cout << "+1 sigma band " << band1sigUp << endl;
cout << "+2 sigma band " << band2sigUp << endl;
// print out the interval on the first Parameter of Interest
cout << "\nobserved 95% upper-limit "<< interval->UpperLimit(*firstPOI) <<endl;
cout << "CLb strict [P(toy>obs|0)] for observed 95% upper-limit "<< CLb <<endl;
cout << "CLb inclusive [P(toy>=obs|0)] for observed 95% upper-limit "<< CLbinclusive <<endl;
delete profile;
delete nll;
}
void fillWorkspace(RooWorkspace& w)
{
// C r e a t e m o d e l
// -----------------------
// Declare observable x
RooRealVar x("x","x",0,10) ;
// Create two Gaussian PDFs g1(x,mean1,sigma) anf g2(x,mean2,sigma) and their parameters
RooRealVar mean("mean","mean of gaussians",5,0,10) ;
RooRealVar sigma1("sigma1","width of gaussians",0.5) ;
RooRealVar sigma2("sigma2","width of gaussians",1) ;
RooGaussian sig1("sig1","Signal component 1",x,mean,sigma1) ;
RooGaussian sig2("sig2","Signal component 2",x,mean,sigma2) ;
// Build Chebychev polynomial p.d.f.
RooRealVar a0("a0","a0",0.5,0.,1.) ;
RooRealVar a1("a1","a1",-0.2,0.,1.) ;
RooChebychev bkg("bkg","Background",x,RooArgSet(a0,a1)) ;
// Sum the signal components into a composite signal p.d.f.
RooRealVar sig1frac("sig1frac","fraction of component 1 in signal",0.8,0.,1.) ;
RooAddPdf sig("sig","Signal",RooArgList(sig1,sig2),sig1frac) ;
// Sum the composite signal and background
RooRealVar bkgfrac("bkgfrac","fraction of background",0.5,0.,1.) ;
RooAddPdf model("model","g1+g2+a",RooArgList(bkg,sig),bkgfrac) ;
// Import model into p.d.f.
w.import(model) ;
// E n c o d e d e f i n i t i o n o f p a r a m e t e r s i n w o r k s p a c e
// ---------------------------------------------------------------------------------------
// Define named sets "parameters" and "observables", which list which variables should be considered
// parameters and observables by the users convention
//
// Variables appearing in sets _must_ live in the workspace already, or the autoImport flag
// of defineSet must be set to import them on the fly. Named sets contain only references
// to the original variables, therefore the value of observables in named sets already
// reflect their 'current' value
RooArgSet* params = (RooArgSet*) model.getParameters(x) ;
w.defineSet("parameters",*params) ;
w.defineSet("observables",x) ;
// E n c o d e r e f e r e n c e v a l u e f o r p a r a m e t e r s i n w o r k s p a c e
// ---------------------------------------------------------------------------------------------------
// Define a parameter 'snapshot' in the p.d.f.
// Unlike a named set, a parameter snapshot stores an independent set of values for
// a given set of variables in the workspace. The values can be stored and reloaded
// into the workspace variable objects using the loadSnapshot() and saveSnapshot()
// methods. A snapshot saves the value of each variable, any errors that are stored
// with it as well as the 'Constant' flag that is used in fits to determine if a
// parameter is kept fixed or not.
// Do a dummy fit to a (supposedly) reference dataset here and store the results
// of that fit into a snapshot
RooDataSet* refData = model.generate(x,10000) ;
model.fitTo(*refData,PrintLevel(-1)) ;
// The kTRUE flag imports the values of the objects in (*params) into the workspace
// If not set, the present values of the workspace parameters objects are stored
w.saveSnapshot("reference_fit",*params,kTRUE) ;
// Make another fit with the signal componentforced to zero
// and save those parameters too
bkgfrac.setVal(1) ;
bkgfrac.setConstant(kTRUE) ;
bkgfrac.removeError() ;
model.fitTo(*refData,PrintLevel(-1)) ;
w.saveSnapshot("reference_fit_bkgonly",*params,kTRUE) ;
}
void OneSidedFrequentistUpperLimitWithBands_intermediate(const char* infile = "",
const char* workspaceName = "combined",
const char* modelConfigName = "ModelConfig",
const char* dataName = "obsData"){
double confidenceLevel=0.95;
// degrade/improve number of pseudo-experiments used to define the confidence belt.
// value of 1 corresponds to default number of toys in the tail, which is 50/(1-confidenceLevel)
double additionalToysFac = 1.;
int nPointsToScan = 30; // number of steps in the parameter of interest
int nToyMC = 100; // number of toys used to define the expected limit and band
TStopwatch t;
t.Start();
/////////////////////////////////////////////////////////////
// First part is just to access a user-defined file
// or create the standard example file if it doesn't exist
////////////////////////////////////////////////////////////
const char* filename = "";
if (!strcmp(infile,""))
filename = "results/example_combined_GaussExample_model.root";
else
filename = infile;
// Check if example input file exists
TFile *file = TFile::Open(filename);
// if input file was specified byt not found, quit
if(!file && strcmp(infile,"")){
cout <<"file not found" << endl;
return;
}
// if default file not found, try to create it
if(!file ){
// Normally this would be run on the command line
cout <<"will run standard hist2workspace example"<<endl;
gROOT->ProcessLine(".! prepareHistFactory .");
gROOT->ProcessLine(".! hist2workspace config/example.xml");
cout <<"\n\n---------------------"<<endl;
cout <<"Done creating example input"<<endl;
cout <<"---------------------\n\n"<<endl;
}
// now try to access the file again
file = TFile::Open(filename);
if(!file){
// if it is still not there, then we can't continue
cout << "Not able to run hist2workspace to create example input" <<endl;
return;
}
/////////////////////////////////////////////////////////////
// Now get the data and workspace
////////////////////////////////////////////////////////////
// get the workspace out of the file
RooWorkspace* w = (RooWorkspace*) file->Get(workspaceName);
if(!w){
cout <<"workspace not found" << endl;
return;
}
// get the modelConfig out of the file
ModelConfig* mc = (ModelConfig*) w->obj(modelConfigName);
// get the modelConfig out of the file
RooAbsData* data = w->data(dataName);
// make sure ingredients are found
if(!data || !mc){
w->Print();
cout << "data or ModelConfig was not found" <<endl;
return;
}
cout << "Found data and ModelConfig:" <<endl;
mc->Print();
/////////////////////////////////////////////////////////////
// Now get the POI for convenience
// you may want to adjust the range of your POI
////////////////////////////////////////////////////////////
RooRealVar* firstPOI = (RooRealVar*) mc->GetParametersOfInterest()->first();
// firstPOI->setMin(0);
// firstPOI->setMax(10);
/////////////////////////////////////////////
// create and use the FeldmanCousins tool
// to find and plot the 95% confidence interval
// on the parameter of interest as specified
// in the model config
// REMEMBER, we will change the test statistic
// so this is NOT a Feldman-Cousins interval
FeldmanCousins fc(*data,*mc);
fc.SetConfidenceLevel(confidenceLevel);
fc.AdditionalNToysFactor(additionalToysFac); // improve sampling that defines confidence belt
// fc.UseAdaptiveSampling(true); // speed it up a bit, but don't use for expectd limits
fc.SetNBins(nPointsToScan); // set how many points per parameter of interest to scan
fc.CreateConfBelt(true); // save the information in the belt for plotting
/////////////////////////////////////////////
// Feldman-Cousins is a unified limit by definition
// but the tool takes care of a few things for us like which values
// of the nuisance parameters should be used to generate toys.
// so let's just change the test statistic and realize this is
// no longer "Feldman-Cousins" but is a fully frequentist Neyman-Construction.
// ProfileLikelihoodTestStatModified onesided(*mc->GetPdf());
// fc.GetTestStatSampler()->SetTestStatistic(&onesided);
// ((ToyMCSampler*) fc.GetTestStatSampler())->SetGenerateBinned(true);
ToyMCSampler* toymcsampler = (ToyMCSampler*) fc.GetTestStatSampler();
ProfileLikelihoodTestStat* testStat = dynamic_cast<ProfileLikelihoodTestStat*>(toymcsampler->GetTestStatistic());
testStat->SetOneSided(true);
// test speedups:
testStat->SetReuseNLL(true);
// toymcsampler->setUseMultiGen(true); // not fully validated
// Since this tool needs to throw toy MC the PDF needs to be
// extended or the tool needs to know how many entries in a dataset
// per pseudo experiment.
// In the 'number counting form' where the entries in the dataset
// are counts, and not values of discriminating variables, the
// datasets typically only have one entry and the PDF is not
// extended.
if(!mc->GetPdf()->canBeExtended()){
if(data->numEntries()==1)
fc.FluctuateNumDataEntries(false);
else
cout <<"Not sure what to do about this model" <<endl;
}
// We can use PROOF to speed things along in parallel
ProofConfig pc(*w, 4, "",false);
if(mc->GetGlobalObservables()){
cout << "will use global observables for unconditional ensemble"<<endl;
mc->GetGlobalObservables()->Print();
toymcsampler->SetGlobalObservables(*mc->GetGlobalObservables());
}
toymcsampler->SetProofConfig(&pc); // enable proof
// Now get the interval
PointSetInterval* interval = fc.GetInterval();
ConfidenceBelt* belt = fc.GetConfidenceBelt();
// print out the iterval on the first Parameter of Interest
cout << "\n95% interval on " <<firstPOI->GetName()<<" is : ["<<
interval->LowerLimit(*firstPOI) << ", "<<
interval->UpperLimit(*firstPOI) <<"] "<<endl;
// get observed UL and value of test statistic evaluated there
RooArgSet tmpPOI(*firstPOI);
double observedUL = interval->UpperLimit(*firstPOI);
firstPOI->setVal(observedUL);
double obsTSatObsUL = fc.GetTestStatSampler()->EvaluateTestStatistic(*data,tmpPOI);
// Ask the calculator which points were scanned
RooDataSet* parameterScan = (RooDataSet*) fc.GetPointsToScan();
RooArgSet* tmpPoint;
// make a histogram of parameter vs. threshold
TH1F* histOfThresholds = new TH1F("histOfThresholds","",
parameterScan->numEntries(),
firstPOI->getMin(),
firstPOI->getMax());
histOfThresholds->GetXaxis()->SetTitle(firstPOI->GetName());
histOfThresholds->GetYaxis()->SetTitle("Threshold");
// loop through the points that were tested and ask confidence belt
// what the upper/lower thresholds were.
// For FeldmanCousins, the lower cut off is always 0
for(Int_t i=0; i<parameterScan->numEntries(); ++i){
tmpPoint = (RooArgSet*) parameterScan->get(i)->clone("temp");
double arMax = belt->GetAcceptanceRegionMax(*tmpPoint);
double poiVal = tmpPoint->getRealValue(firstPOI->GetName()) ;
histOfThresholds->Fill(poiVal,arMax);
}
TCanvas* c1 = new TCanvas();
c1->Divide(2);
c1->cd(1);
histOfThresholds->SetMinimum(0);
histOfThresholds->Draw();
c1->cd(2);
/////////////////////////////////////////////////////////////
// Now we generate the expected bands and power-constriant
////////////////////////////////////////////////////////////
// First: find parameter point for mu=0, with conditional MLEs for nuisance parameters
RooAbsReal* nll = mc->GetPdf()->createNLL(*data);
RooAbsReal* profile = nll->createProfile(*mc->GetParametersOfInterest());
firstPOI->setVal(0.);
profile->getVal(); // this will do fit and set nuisance parameters to profiled values
RooArgSet* poiAndNuisance = new RooArgSet();
if(mc->GetNuisanceParameters())
poiAndNuisance->add(*mc->GetNuisanceParameters());
poiAndNuisance->add(*mc->GetParametersOfInterest());
w->saveSnapshot("paramsToGenerateData",*poiAndNuisance);
RooArgSet* paramsToGenerateData = (RooArgSet*) poiAndNuisance->snapshot();
cout << "\nWill use these parameter points to generate pseudo data for bkg only" << endl;
paramsToGenerateData->Print("v");
double CLb=0;
double CLbinclusive=0;
// Now we generate background only and find distribution of upper limits
TH1F* histOfUL = new TH1F("histOfUL","",100,0,firstPOI->getMax());
histOfUL->GetXaxis()->SetTitle("Upper Limit (background only)");
histOfUL->GetYaxis()->SetTitle("Entries");
for(int imc=0; imc<nToyMC; ++imc){
// set parameters back to values for generating pseudo data
w->loadSnapshot("paramsToGenerateData");
// in 5.30 there is a nicer way to generate toy data & randomize global obs
RooAbsData* toyData = toymcsampler->GenerateToyData(*paramsToGenerateData);
// get test stat at observed UL in observed data
firstPOI->setVal(observedUL);
double toyTSatObsUL = fc.GetTestStatSampler()->EvaluateTestStatistic(*toyData,tmpPOI);
// toyData->get()->Print("v");
// cout <<"obsTSatObsUL " <<obsTSatObsUL << "toyTS " << toyTSatObsUL << endl;
if(obsTSatObsUL < toyTSatObsUL) // (should be checked)
CLb+= (1.)/nToyMC;
if(obsTSatObsUL <= toyTSatObsUL) // (should be checked)
CLbinclusive+= (1.)/nToyMC;
// loop over points in belt to find upper limit for this toy data
double thisUL = 0;
for(Int_t i=0; i<parameterScan->numEntries(); ++i){
tmpPoint = (RooArgSet*) parameterScan->get(i)->clone("temp");
double arMax = belt->GetAcceptanceRegionMax(*tmpPoint);
firstPOI->setVal( tmpPoint->getRealValue(firstPOI->GetName()) );
double thisTS = fc.GetTestStatSampler()->EvaluateTestStatistic(*toyData,tmpPOI);
if(thisTS<=arMax){
thisUL = firstPOI->getVal();
} else{
break;
}
}
histOfUL->Fill(thisUL);
delete toyData;
}
histOfUL->Draw();
c1->SaveAs("one-sided_upper_limit_output.pdf");
// if you want to see a plot of the sampling distribution for a particular scan point:
// Now find bands and power constraint
Double_t* bins = histOfUL->GetIntegral();
TH1F* cumulative = (TH1F*) histOfUL->Clone("cumulative");
cumulative->SetContent(bins);
double band2sigDown=0, band1sigDown=0, bandMedian=0, band1sigUp=0,band2sigUp=0;
for(int i=1; i<=cumulative->GetNbinsX(); ++i){
if(bins[i]<RooStats::SignificanceToPValue(2))
band2sigDown=cumulative->GetBinCenter(i);
if(bins[i]<RooStats::SignificanceToPValue(1))
band1sigDown=cumulative->GetBinCenter(i);
if(bins[i]<0.5)
bandMedian=cumulative->GetBinCenter(i);
if(bins[i]<RooStats::SignificanceToPValue(-1))
band1sigUp=cumulative->GetBinCenter(i);
if(bins[i]<RooStats::SignificanceToPValue(-2))
band2sigUp=cumulative->GetBinCenter(i);
}
t.Stop();
t.Print();
cout << "-2 sigma band " << band2sigDown << endl;
cout << "-1 sigma band " << band1sigDown << endl;
cout << "median of band " << bandMedian << " [Power Constriant)]" << endl;
cout << "+1 sigma band " << band1sigUp << endl;
cout << "+2 sigma band " << band2sigUp << endl;
// print out the iterval on the first Parameter of Interest
cout << "\nobserved 95% upper-limit "<< interval->UpperLimit(*firstPOI) <<endl;
cout << "CLb strict [P(toy>obs|0)] for observed 95% upper-limit "<< CLb <<endl;
cout << "CLb inclusive [P(toy>=obs|0)] for observed 95% upper-limit "<< CLbinclusive <<endl;
delete profile;
delete nll;
}
bool fitCharmoniaMassModel( RooWorkspace& myws, // Local Workspace
const RooWorkspace& inputWorkspace, // Workspace with all the input RooDatasets
struct KinCuts& cut, // Variable containing all kinematic cuts
map<string, string>& parIni, // Variable containing all initial parameters
struct InputOpt& opt, // Variable with run information (kept for legacy purpose)
string outputDir, // Path to output directory
// Select the type of datasets to fit
string DSTAG, // Specifies the type of datasets: i.e, DATA, MCJPSINP, ...
bool isPbPb = false, // isPbPb = false for pp, true for PbPb
bool importDS = true, // Select if the dataset is imported in the local workspace
// Select the type of object to fit
bool incJpsi = true, // Includes Jpsi model
bool incPsi2S = true, // Includes Psi(2S) model
bool incBkg = true, // Includes Background model
// Select the fitting options
bool doFit = true, // Flag to indicate if we want to perform the fit
bool cutCtau = false, // Apply prompt ctau cuts
bool doConstrFit = false, // Do constrained fit
bool doSimulFit = false, // Do simultaneous fit
bool wantPureSMC = false, // Flag to indicate if we want to fit pure signal MC
const char* applyCorr ="", // Flag to indicate if we want corrected dataset and which correction
uint loadFitResult = false, // Load previous fit results
string inputFitDir = "", // Location of the fit results
int numCores = 2, // Number of cores used for fitting
// Select the drawing options
bool setLogScale = true, // Draw plot with log scale
bool incSS = false, // Include Same Sign data
bool zoomPsi = false, // Zoom Psi(2S) peak on extra pad
double binWidth = 0.05, // Bin width used for plotting
bool getMeanPT = false // Compute the mean PT (NEED TO FIX)
)
{
if (DSTAG.find("_")!=std::string::npos) DSTAG.erase(DSTAG.find("_"));
// Check if input dataset is MC
bool isMC = false;
if (DSTAG.find("MC")!=std::string::npos) {
if (incJpsi && incPsi2S) {
cout << "[ERROR] We can only fit one type of signal using MC" << endl; return false;
}
isMC = true;
}
wantPureSMC = (isMC && wantPureSMC);
bool cutSideBand = (incBkg && (!incPsi2S && !incJpsi));
bool applyWeight_Corr = ( strcmp(applyCorr,"") );
// Define the mass range
setMassCutParameters(cut, incJpsi, incPsi2S, isMC);
parIni["invMassNorm"] = Form("RooFormulaVar::%s('( -1.0 + 2.0*( @0 - @1 )/( @2 - @1) )', {%s, mMin[%.6f], mMax[%.6f]})", "invMassNorm", "invMass", cut.dMuon.M.Min, cut.dMuon.M.Max );
// Apply the ctau cuts to reject non-prompt charmonia
if (cutCtau) { setCtauCuts(cut, isPbPb); }
string COLL = (isPbPb ? "PbPb" : "PP" );
string plotLabelPbPb, plotLabelPP;
if (doSimulFit || !isPbPb) {
// Set models based on initial parameters
struct OniaModel model;
if (!setMassModel(model, parIni, false, incJpsi, incPsi2S, incBkg)) { return false; }
// Import the local datasets
double numEntries = 1000000;
string label = ((DSTAG.find("PP")!=std::string::npos) ? DSTAG.c_str() : Form("%s_%s", DSTAG.c_str(), "PP"));
if (wantPureSMC) label += "_NoBkg";
if (applyWeight_Corr) label += Form("_%s",applyCorr);
string dsName = Form("dOS_%s", label.c_str());
if (importDS) {
if ( !(myws.data(dsName.c_str())) ) {
int importID = importDataset(myws, inputWorkspace, cut, label, cutSideBand);
if (importID<0) { return false; }
else if (importID==0) { doFit = false; }
}
numEntries = myws.data(dsName.c_str())->sumEntries(); if (numEntries<=0) { doFit = false; }
}
else if (doFit && !(myws.data(dsName.c_str()))) { cout << "[ERROR] No local dataset was found to perform the fit!" << endl; return false; }
if (myws.data(dsName.c_str())) numEntries = myws.data(dsName.c_str())->sumEntries();
// Set global parameters
setMassGlobalParameterRange(myws, parIni, cut, incJpsi, incPsi2S, incBkg, wantPureSMC);
// Build the Fit Model
if (!buildCharmoniaMassModel(myws, model.PP, parIni, false, doConstrFit, doSimulFit, incBkg, incJpsi, incPsi2S, numEntries)) { return false; }
// Define plot names
if (incJpsi) { plotLabelPP += Form("_Jpsi_%s", parIni["Model_Jpsi_PP"].c_str()); }
if (incPsi2S) { plotLabelPP += Form("_Psi2S_%s", parIni["Model_Psi2S_PP"].c_str()); }
if (incBkg) { plotLabelPP += Form("_Bkg_%s", parIni["Model_Bkg_PP"].c_str()); }
if (wantPureSMC) plotLabelPP +="_NoBkg";
if (applyWeight_Corr) plotLabelPP +=Form("_%s",applyCorr);
}
if (doSimulFit || isPbPb) {
// Set models based on initial parameters
struct OniaModel model;
if (!setMassModel(model, parIni, true, incJpsi, incPsi2S, incBkg)) { return false; }
// Import the local datasets
double numEntries = 1000000;
string label = ((DSTAG.find("PbPb")!=std::string::npos) ? DSTAG.c_str() : Form("%s_%s", DSTAG.c_str(), "PbPb"));
if (wantPureSMC) label += "_NoBkg";
if (applyWeight_Corr) label += Form("_%s",applyCorr);
string dsName = Form("dOS_%s", label.c_str());
if (importDS) {
if ( !(myws.data(dsName.c_str())) ) {
int importID = importDataset(myws, inputWorkspace, cut, label, cutSideBand);
if (importID<0) { return false; }
else if (importID==0) { doFit = false; }
}
numEntries = myws.data(dsName.c_str())->sumEntries(); if (numEntries<=0) { doFit = false; }
}
else if (doFit && !(myws.data(dsName.c_str()))) { cout << "[ERROR] No local dataset was found to perform the fit!" << endl; return false; }
if (myws.data(dsName.c_str())) numEntries = myws.data(dsName.c_str())->sumEntries();
// Set global parameters
setMassGlobalParameterRange(myws, parIni, cut, incJpsi, incPsi2S, incBkg, wantPureSMC);
// Build the Fit Model
if (!buildCharmoniaMassModel(myws, model.PbPb, parIni, true, doConstrFit, doSimulFit, incBkg, incJpsi, incPsi2S, numEntries)) { return false; }
// Define plot names
if (incJpsi) { plotLabelPbPb += Form("_Jpsi_%s", parIni["Model_Jpsi_PbPb"].c_str()); }
if (incPsi2S) { plotLabelPbPb += Form("_Psi2S_%s", parIni["Model_Psi2S_PbPb"].c_str()); }
if (incBkg) { plotLabelPbPb += Form("_Bkg_%s", parIni["Model_Bkg_PbPb"].c_str()); }
if (wantPureSMC) plotLabelPbPb += "_NoBkg";
if (applyWeight_Corr) plotLabelPbPb += Form("_%s",applyCorr);
}
if (doSimulFit) {
// Create the combided datasets
RooCategory* sample = new RooCategory("sample","sample"); sample->defineType("PbPb"); sample->defineType("PP");
RooDataSet* combData = new RooDataSet("combData","combined data", *myws.var("invMass"), Index(*sample),
Import("PbPb", *((RooDataSet*)myws.data("dOS_DATA_PbPb"))),
Import("PP", *((RooDataSet*)myws.data("dOS_DATA_PP")))
);
myws.import(*sample);
// Create the combided models
RooSimultaneous* simPdf = new RooSimultaneous("simPdf", "simultaneous pdf", *sample);
simPdf->addPdf(*myws.pdf("pdfMASS_Tot_PbPb"), "PbPb"); simPdf->addPdf(*myws.pdf("pdfMASS_Tot_PP"), "PP");
myws.import(*simPdf);
// check if we have already done this fit. If yes, do nothing and return true.
string FileName = "";
setMassFileName(FileName, (inputFitDir=="" ? outputDir : inputFitDir), DSTAG, (plotLabelPP + plotLabelPbPb), cut, isPbPb, cutSideBand, doSimulFit);
if (gSystem->AccessPathName(FileName.c_str()) && inputFitDir!="") {
cout << "[WARNING] User Input File : " << FileName << " was not found!" << endl;
if (loadFitResult) return false;
setMassFileName(FileName, outputDir, DSTAG, (plotLabelPP + plotLabelPbPb), cut, isPbPb, cutSideBand, doSimulFit);
}
bool found = true; bool skipFit = !doFit;
RooArgSet *newpars = myws.pdf("simPdf")->getParameters(*(myws.var("invMass")));
myws.saveSnapshot("simPdf_parIni", *newpars, kTRUE);
found = found && isFitAlreadyFound(newpars, FileName, "simPdf");
if (loadFitResult) {
if ( loadPreviousFitResult(myws, FileName, DSTAG, false, (!isMC && !cutSideBand && loadFitResult==1), loadFitResult==1) ) { skipFit = true; } else { skipFit = false; }
if ( loadPreviousFitResult(myws, FileName, DSTAG, true, (!isMC && !cutSideBand && loadFitResult==1), loadFitResult==1) ) { skipFit = true; } else { skipFit = false; }
if (skipFit) { cout << "[INFO] This simultaneous mass fit was already done, so I'll load the fit results." << endl; }
myws.saveSnapshot("simPdf_parLoad", *newpars, kTRUE);
} else if (found) {
cout << "[INFO] This simultaneous mass fit was already done, so I'll just go to the next one." << endl;
return true;
}
// Do the simultaneous fit
if (skipFit==false) {
RooFitResult* fitResult = simPdf->fitTo(*combData, Offset(kTRUE), Extended(kTRUE), NumCPU(numCores), Range("MassWindow"), Save()); //, Minimizer("Minuit2","Migrad")
fitResult->Print("v");
myws.import(*fitResult, "fitResult_simPdf");
// Create the output files
int nBins = min(int( round((cut.dMuon.M.Max - cut.dMuon.M.Min)/binWidth) ), 1000);
drawMassPlot(myws, outputDir, opt, cut, parIni, plotLabelPP, DSTAG, false, incJpsi, incPsi2S, incBkg, cutCtau, doSimulFit, false, setLogScale, incSS, zoomPsi, nBins, getMeanPT);
drawMassPlot(myws, outputDir, opt, cut, parIni, plotLabelPbPb, DSTAG, true, incJpsi, incPsi2S, incBkg, cutCtau, doSimulFit, false, setLogScale, incSS, zoomPsi, nBins, getMeanPT);
// Save the results
string FileName = ""; setMassFileName(FileName, outputDir, DSTAG, (plotLabelPP + plotLabelPbPb), cut, isPbPb, cutSideBand, doSimulFit);
myws.saveSnapshot("simPdf_parFit", *newpars, kTRUE);
saveWorkSpace(myws, Form("%smass%s/%s/result", outputDir.c_str(), (cutSideBand?"SB":""), DSTAG.c_str()), FileName);
// Delete the objects used during the simultaneous fit
delete sample; delete combData; delete simPdf;
}
}
else {
// Define pdf and plot names
string pdfName = Form("pdfMASS_Tot_%s", COLL.c_str());
string plotLabel = (isPbPb ? plotLabelPbPb : plotLabelPP);
// Import the local datasets
string label = ((DSTAG.find(COLL.c_str())!=std::string::npos) ? DSTAG.c_str() : Form("%s_%s", DSTAG.c_str(), COLL.c_str()));
if (wantPureSMC) label += "_NoBkg";
if (applyWeight_Corr) label += Form("_%s",applyCorr);
string dsName = Form("dOS_%s", label.c_str());
// check if we have already done this fit. If yes, do nothing and return true.
string FileName = "";
setMassFileName(FileName, (inputFitDir=="" ? outputDir : inputFitDir), DSTAG, plotLabel, cut, isPbPb, cutSideBand);
if (gSystem->AccessPathName(FileName.c_str()) && inputFitDir!="") {
cout << "[WARNING] User Input File : " << FileName << " was not found!" << endl;
if (loadFitResult) return false;
setMassFileName(FileName, outputDir, DSTAG, plotLabel, cut, isPbPb, cutSideBand);
}
bool found = true; bool skipFit = !doFit;
RooArgSet *newpars = myws.pdf(pdfName.c_str())->getParameters(*(myws.var("invMass")));
found = found && isFitAlreadyFound(newpars, FileName, pdfName.c_str());
if (loadFitResult) {
if ( loadPreviousFitResult(myws, FileName, DSTAG, isPbPb, (!isMC && !cutSideBand && loadFitResult==1), loadFitResult==1) ) { skipFit = true; } else { skipFit = false; }
if (skipFit) { cout << "[INFO] This mass fit was already done, so I'll load the fit results." << endl; }
myws.saveSnapshot(Form("%s_parLoad", pdfName.c_str()), *newpars, kTRUE);
} else if (found) {
cout << "[INFO] This mass fit was already done, so I'll just go to the next one." << endl;
return true;
}
// Fit the Datasets
if (skipFit==false) {
bool isWeighted = myws.data(dsName.c_str())->isWeighted();
RooFitResult* fitResult(0x0);
if (doConstrFit)
{
cout << "[INFO] Performing constrained fit" << endl;
if (isPbPb) {
cout << "[INFO] Constrained variables: alpha, n, ratio of sigmas" << endl;
fitResult = myws.pdf(pdfName.c_str())->fitTo(*myws.data(dsName.c_str()), Extended(kTRUE), SumW2Error(isWeighted), Range(cutSideBand ? parIni["BkgMassRange_FULL_Label"].c_str() : "MassWindow"), ExternalConstraints(RooArgSet(*(myws.pdf("sigmaAlphaConstr")),*(myws.pdf("sigmaNConstr")),*(myws.pdf("sigmaRSigmaConstr")))), NumCPU(numCores), Save());
}
else {
cout << "[INFO] Constrained variables: alpha, n, ratio of sigmas" << endl;
fitResult = myws.pdf(pdfName.c_str())->fitTo(*myws.data(dsName.c_str()), Extended(kTRUE), SumW2Error(isWeighted), Range(cutSideBand ? parIni["BkgMassRange_FULL_Label"].c_str() : "MassWindow"), ExternalConstraints(RooArgSet(*(myws.pdf("sigmaAlphaConstr")),*(myws.pdf("sigmaNConstr")))), NumCPU(numCores), Save());
}
}
else
{
fitResult = myws.pdf(pdfName.c_str())->fitTo(*myws.data(dsName.c_str()), Extended(kTRUE), SumW2Error(isWeighted), Range(cutSideBand ? parIni["BkgMassRange_FULL_Label"].c_str() : "MassWindow"), NumCPU(numCores), Save());
}
fitResult->Print("v");
myws.import(*fitResult, Form("fitResult_%s", pdfName.c_str()));
// Create the output files
int nBins = min(int( round((cut.dMuon.M.Max - cut.dMuon.M.Min)/binWidth) ), 1000);
drawMassPlot(myws, outputDir, opt, cut, parIni, plotLabel, DSTAG, isPbPb, incJpsi, incPsi2S, incBkg, cutCtau, doSimulFit, wantPureSMC, setLogScale, incSS, zoomPsi, nBins, getMeanPT);
// Save the results
string FileName = ""; setMassFileName(FileName, outputDir, DSTAG, plotLabel, cut, isPbPb, cutSideBand);
myws.saveSnapshot(Form("%s_parFit", pdfName.c_str()), *newpars, kTRUE);
saveWorkSpace(myws, Form("%smass%s/%s/result", outputDir.c_str(), (cutSideBand?"SB":""), DSTAG.c_str()), FileName);
}
}
return true;
};
Int_t Tprime::SetParameterPoints( std::string sbModelName,
std::string bModelName ) {
//
// Fit the data with S+B model.
// Make a snapshot of the S+B parameter point.
// Profile with POI=0.
// Make a snapshot of the B parameter point
// (B model is the S+B model with POI=0
//
Double_t poi_value_for_b_model = 0.0;
// get S+B model config from workspace
RooStats::ModelConfig * pSbModel = (RooStats::ModelConfig *)pWs->obj(sbModelName.c_str());
pSbModel->SetWorkspace(*pWs);
// get parameter of interest set
const RooArgSet * poi = pSbModel->GetParametersOfInterest();
// get B model config from workspace
RooStats::ModelConfig * pBModel = (RooStats::ModelConfig *)pWs->obj(bModelName.c_str());
pBModel->SetWorkspace(*pWs);
// make sure that data has been loaded
if (!data) return -1;
// find parameter point for global maximum with the S+B model,
// with conditional MLEs for nuisance parameters
// and save the parameter point snapshot in the Workspace
RooAbsReal * nll = pSbModel->GetPdf()->createNLL(*data);
RooAbsReal * profile = nll->createProfile(RooArgSet());
profile->getVal(); // this will do fit and set POI and nuisance parameters to fitted values
RooArgSet * poiAndNuisance = new RooArgSet();
if(pSbModel->GetNuisanceParameters())
poiAndNuisance->add(*pSbModel->GetNuisanceParameters());
poiAndNuisance->add(*pSbModel->GetParametersOfInterest());
pWs->defineSet("SPlusBModelParameters", *poiAndNuisance);
pWs->saveSnapshot("SPlusBFitParameters",*poiAndNuisance);
pSbModel->SetSnapshot(*poi);
RooArgSet * sbModelFitParams = (RooArgSet *)poiAndNuisance->snapshot();
cout << "\nWill save these parameter points that correspond to the fit to data" << endl;
sbModelFitParams->Print("v");
delete profile;
delete nll;
delete poiAndNuisance;
delete sbModelFitParams;
//
// Find a parameter point for generating pseudo-data
// with the background-only data.
// Save the parameter point snapshot in the Workspace
nll = pBModel->GetPdf()->createNLL(*data);
profile = nll->createProfile(*poi);
((RooRealVar *)poi->first())->setVal(poi_value_for_b_model);
profile->getVal(); // this will do fit and set nuisance parameters to profiled values
poiAndNuisance = new RooArgSet();
if(pBModel->GetNuisanceParameters())
poiAndNuisance->add(*pBModel->GetNuisanceParameters());
poiAndNuisance->add(*pBModel->GetParametersOfInterest());
pWs->defineSet("parameterPointToGenerateData", *poiAndNuisance);
pWs->saveSnapshot("parametersToGenerateData",*poiAndNuisance);
pBModel->SetSnapshot(*poi);
RooArgSet * paramsToGenerateData = (RooArgSet *)poiAndNuisance->snapshot();
cout << "\nShould use these parameter points to generate pseudo data for bkg only" << endl;
paramsToGenerateData->Print("v");
delete profile;
delete nll;
delete poiAndNuisance;
delete paramsToGenerateData;
return 0;
}
