Logo ROOT  
Reference Guide
 
Loading...
Searching...
No Matches
RooStats::AsymptoticCalculator Class Reference

Hypothesis Test Calculator based on the asymptotic formulae for the profile likelihood ratio.

It performs hypothesis tests using the asymptotic formula for the profile likelihood, and uses the Asimov data set to compute expected significances or limits.

See G. Cowan, K. Cranmer, E. Gross and O. Vitells: Asymptotic formulae for likelihood- based tests of new physics. Eur. Phys. J., C71:1–19, 2011. It provides methods to perform hypothesis tests using the likelihood function, and computes the \(p\)-values for the null and the alternate hypothesis using the asymptotic formulae for the profile likelihood ratio described in the given paper.

The calculator provides methods to produce the Asimov dataset, i.e. a dataset generated where the observed values are equal to the expected ones. The Asimov data set is then used to compute the observed asymptotic \(p\)-value for the alternate hypothesis and the asymptotic expected \(p\)-values.

The asymptotic formulae are valid only for one POI (parameter of interest). So the calculator works only for one-dimensional (one POI) models. If more than one POI exists, only the first one is used.

The calculator can generate Asimov datasets from two kinds of PDFs:

  • "Counting" distributions: RooPoisson, RooGaussian, or products of RooPoissons.
  • Extended, i.e. number of events can be read off from extended likelihood term.

Definition at line 27 of file AsymptoticCalculator.h.

Public Member Functions

 AsymptoticCalculator (RooAbsData &data, const ModelConfig &altModel, const ModelConfig &nullModel, bool nominalAsimov=false)
 constructor for asymptotic calculator from Data set and ModelConfig
 
 ~AsymptoticCalculator () override
 
const RooArgSetGetBestFitParams () const
 return best fit value for all parameters
 
const RooArgSetGetBestFitPoi () const
 return snapshot of the best fit parameter
 
HypoTestResultGetHypoTest () const override
 re-implement HypoTest computation using the asymptotic
 
const RooRealVarGetMuHat () const
 return best fit parameter (firs of poi)
 
bool Initialize () const
 initialize the calculator by performing a global fit and make the Asimov data set
 
TClassIsA () const override
 
bool IsOneSidedDiscovery () const
 
bool IsTwoSided () const
 
void SetAlternateModel (const ModelConfig &altModel) override
 Set the model for the alternate hypothesis.
 
void SetData (RooAbsData &data) override
 Set the DataSet.
 
void SetNullModel (const ModelConfig &nullModel) override
 re-implementation of setters since they are needed to re-initialize the calculator
 
void SetOneSided (bool on)
 set test statistic for one sided (upper limits)
 
void SetOneSidedDiscovery (bool on)
 set the test statistics for one-sided discovery
 
void SetQTilde (bool on)
 set using of qtilde, by default is controlled if RoORealVar is limited or not
 
void SetTwoSided ()
 set the test statistics for two sided (in case of upper limits for discovery does not make really sense)
 
void Streamer (TBuffer &) override
 
void StreamerNVirtual (TBuffer &ClassDef_StreamerNVirtual_b)
 
- Public Member Functions inherited from RooStats::HypoTestCalculatorGeneric
 HypoTestCalculatorGeneric (const RooAbsData &data, const ModelConfig &altModel, const ModelConfig &nullModel, TestStatSampler *sampler=nullptr)
 Constructor.
 
 ~HypoTestCalculatorGeneric () override
 
const ModelConfigGetAlternateModel (void) const
 
const RooAbsDataGetData (void) const
 
virtual const RooArgSetGetFitInfo () const
 
HypoTestResultGetHypoTest () const override
 inherited methods from HypoTestCalculator interface
 
const ModelConfigGetNullModel (void) const
 
TestStatSamplerGetTestStatSampler (void) const
 Returns instance of TestStatSampler.
 
TClassIsA () const override
 
void SetAlternateModel (const ModelConfig &altModel) override
 Set the model for the alternate hypothesis (S+B)
 
void SetData (RooAbsData &data) override
 Set the DataSet.
 
void SetNullModel (const ModelConfig &nullModel) override
 set the model for the null hypothesis (only B)
 
void Streamer (TBuffer &) override
 
void StreamerNVirtual (TBuffer &ClassDef_StreamerNVirtual_b)
 
void UseSameAltToys ()
 Set this for re-using always the same toys for alternate hypothesis in case of calls at different null parameter points This is useful to get more stable bands when running the HypoTest inversion.
 
- Public Member Functions inherited from RooStats::HypoTestCalculator
virtual ~HypoTestCalculator ()
 
virtual void SetCommonModel (const ModelConfig &model)
 Set a common model for both the null and alternate, add to the workspace if not already there.
 
void StreamerNVirtual (TBuffer &ClassDef_StreamerNVirtual_b)
 

Static Public Member Functions

static TClassClass ()
 
static const char * Class_Name ()
 
static constexpr Version_t Class_Version ()
 
static const char * DeclFileName ()
 
static RooAbsDataGenerateAsimovData (const RooAbsPdf &pdf, const RooArgSet &observables)
 generate the asimov data for the observables (not the global ones) need to deal with the case of a sim pdf
 
static double GetExpectedPValues (double pnull, double palt, double nsigma, bool usecls, bool oneSided=true)
 function given the null and the alt p value - return the expected one given the N - sigma value
 
static RooAbsDataMakeAsimovData (const ModelConfig &model, const RooArgSet &allParamValues, RooArgSet &globObs)
 Make a nominal Asimov data set from a model.
 
static RooAbsDataMakeAsimovData (RooAbsData &data, const ModelConfig &model, const RooArgSet &poiValues, RooArgSet &globObs, const RooArgSet *genPoiValues=nullptr)
 Make Asimov data.
 
static void SetPrintLevel (int level)
 set print level (static function)
 
- Static Public Member Functions inherited from RooStats::HypoTestCalculatorGeneric
static TClassClass ()
 
static const char * Class_Name ()
 
static constexpr Version_t Class_Version ()
 
static const char * DeclFileName ()
 
- Static Public Member Functions inherited from RooStats::HypoTestCalculator
static TClassClass ()
 
static const char * Class_Name ()
 
static constexpr Version_t Class_Version ()
 
static const char * DeclFileName ()
 

Static Protected Member Functions

static double EvaluateNLL (RooStats::ModelConfig const &modelConfig, RooAbsData &data, const RooArgSet *poiSet=nullptr)
 
static void FillBins (const RooAbsPdf &pdf, const RooArgList &obs, RooAbsData &data, int &index, double &binVolume, int &ibin)
 fill bins by looping recursively on observables
 
static RooAbsDataGenerateAsimovDataSinglePdf (const RooAbsPdf &pdf, const RooArgSet &obs, const RooRealVar &weightVar, RooCategory *channelCat=nullptr)
 Compute the asimov data set for an observable of a pdf.
 
static RooAbsDataGenerateCountingAsimovData (RooAbsPdf &pdf, const RooArgSet &obs, const RooRealVar &weightVar, RooCategory *channelCat=nullptr)
 Generate counting Asimov data for the case when the pdf cannot be extended.
 
static bool SetObsToExpected (RooAbsPdf &pdf, const RooArgSet &obs)
 set observed value to the expected one works for Gaussian, Poisson or LogNormal assumes mean parameter value is the argument not constant and not depending on observables (if more than two arguments are not constant will use first one but print a warning !) need to iterate on the components of the Poisson to get n and nu (nu can be a RooAbsReal) (code from G.
 
static bool SetObsToExpected (RooProdPdf &prod, const RooArgSet &obs)
 Inpspect a product pdf to find all the Poisson or Gaussian parts to set the observed values to expected ones.
 

Private Attributes

RooAbsDatafAsimovData
 asimov data set
 
RooArgSet fAsimovGlobObs
 snapshot of Asimov global observables
 
RooArgSet fBestFitParams
 snapshot of all best fitted Parameter values
 
RooArgSet fBestFitPoi
 snapshot of best fitted POI values
 
bool fIsInitialized
 ! flag to check if calculator is initialized
 
double fNLLAsimov
 
double fNLLObs
 
bool fNominalAsimov
 make Asimov at nominal parameter values
 
bool fOneSided
 for one sided PL test statistic (upper limits)
 
bool fOneSidedDiscovery
 for one sided PL test statistic (for discovery)
 
int fUseQTilde
 flag to indicate if using qtilde or not (-1 (default based on RooRealVar)), 0 false, 1 (true)
 

Static Private Attributes

static int fgPrintLevel = 1
 control print level (0 minimal, 1 normal, 2 debug)
 

Additional Inherited Members

- Protected Member Functions inherited from RooStats::HypoTestCalculatorGeneric
virtual int CheckHook (void) const
 
virtual void PostHook () const
 
virtual int PreAltHook (RooArgSet *, double) const
 
virtual void PreHook () const
 
virtual int PreNullHook (RooArgSet *, double) const
 
- Protected Attributes inherited from RooStats::HypoTestCalculatorGeneric
const ModelConfigfAltModel
 
unsigned int fAltToysSeed
 
const RooAbsDatafData
 
TestStatSamplerfDefaultSampler
 
TestStatisticfDefaultTestStat
 
const ModelConfigfNullModel
 
TestStatSamplerfTestStatSampler
 

#include <RooStats/AsymptoticCalculator.h>

Inheritance diagram for RooStats::AsymptoticCalculator:
[legend]

Constructor & Destructor Documentation

◆ AsymptoticCalculator()

AsymptoticCalculator::AsymptoticCalculator ( RooAbsData data,
const ModelConfig altModel,
const ModelConfig nullModel,
bool  nominalAsimov = false 
)

constructor for asymptotic calculator from Data set and ModelConfig

Definition at line 93 of file AsymptoticCalculator.cxx.

◆ ~AsymptoticCalculator()

RooStats::AsymptoticCalculator::~AsymptoticCalculator ( )
inlineoverride

Definition at line 37 of file AsymptoticCalculator.h.

Member Function Documentation

◆ Class()

static TClass * RooStats::AsymptoticCalculator::Class ( )
static
Returns
TClass describing this class

◆ Class_Name()

static const char * RooStats::AsymptoticCalculator::Class_Name ( )
static
Returns
Name of this class

◆ Class_Version()

static constexpr Version_t RooStats::AsymptoticCalculator::Class_Version ( )
inlinestaticconstexpr
Returns
Version of this class

Definition at line 125 of file AsymptoticCalculator.h.

◆ DeclFileName()

static const char * RooStats::AsymptoticCalculator::DeclFileName ( )
inlinestatic
Returns
Name of the file containing the class declaration

Definition at line 125 of file AsymptoticCalculator.h.

◆ EvaluateNLL()

double AsymptoticCalculator::EvaluateNLL ( RooStats::ModelConfig const &  modelConfig,
RooAbsData data,
const RooArgSet poiSet = nullptr 
)
staticprotected

Definition at line 289 of file AsymptoticCalculator.cxx.

◆ FillBins()

void AsymptoticCalculator::FillBins ( const RooAbsPdf pdf,
const RooArgList obs,
RooAbsData data,
int index,
double binVolume,
int ibin 
)
staticprotected

fill bins by looping recursively on observables

Definition at line 843 of file AsymptoticCalculator.cxx.

◆ GenerateAsimovData()

RooAbsData * AsymptoticCalculator::GenerateAsimovData ( const RooAbsPdf pdf,
const RooArgSet observables 
)
static

generate the asimov data for the observables (not the global ones) need to deal with the case of a sim pdf

Definition at line 1126 of file AsymptoticCalculator.cxx.

◆ GenerateAsimovDataSinglePdf()

RooAbsData * AsymptoticCalculator::GenerateAsimovDataSinglePdf ( const RooAbsPdf pdf,
const RooArgSet allobs,
const RooRealVar weightVar,
RooCategory channelCat = nullptr 
)
staticprotected

Compute the asimov data set for an observable of a pdf.

It generates binned data following the binning of the observables.

Definition at line 1047 of file AsymptoticCalculator.cxx.

◆ GenerateCountingAsimovData()

RooAbsData * AsymptoticCalculator::GenerateCountingAsimovData ( RooAbsPdf pdf,
const RooArgSet observables,
const RooRealVar weightVar,
RooCategory channelCat = nullptr 
)
staticprotected

Generate counting Asimov data for the case when the pdf cannot be extended.

This function assumes that the pdf is a RooPoisson or can be decomposed in a product of RooPoisson, or is a RooGaussian. Otherwise, we cannot know how to make the Asimov data sets.

Definition at line 1008 of file AsymptoticCalculator.cxx.

◆ GetBestFitParams()

const RooArgSet & RooStats::AsymptoticCalculator::GetBestFitParams ( ) const
inline

return best fit value for all parameters

Definition at line 97 of file AsymptoticCalculator.h.

◆ GetBestFitPoi()

const RooArgSet & RooStats::AsymptoticCalculator::GetBestFitPoi ( ) const
inline

return snapshot of the best fit parameter

Definition at line 93 of file AsymptoticCalculator.h.

◆ GetExpectedPValues()

double AsymptoticCalculator::GetExpectedPValues ( double  pnull,
double  palt,
double  nsigma,
bool  usecls,
bool  oneSided = true 
)
static

function given the null and the alt p value - return the expected one given the N - sigma value

Definition at line 794 of file AsymptoticCalculator.cxx.

◆ GetHypoTest()

HypoTestResult * AsymptoticCalculator::GetHypoTest ( ) const
overridevirtual

re-implement HypoTest computation using the asymptotic

It performs an hypothesis tests using the likelihood function and computes the p values for the null and the alternate using the asymptotic formulae for the profile likelihood ratio.

See G. Cowan, K. Cranmer, E. Gross and O. Vitells. Asymptotic formulae for likelihood- based tests of new physics. Eur. Phys. J., C71:1–19, 2011. The formulae are valid only for one POI. If more than one POI exists consider as POI only the first one

Implements RooStats::HypoTestCalculator.

Definition at line 463 of file AsymptoticCalculator.cxx.

◆ GetMuHat()

const RooRealVar * RooStats::AsymptoticCalculator::GetMuHat ( ) const
inline

return best fit parameter (firs of poi)

Definition at line 95 of file AsymptoticCalculator.h.

◆ Initialize()

bool AsymptoticCalculator::Initialize ( ) const

initialize the calculator by performing a global fit and make the Asimov data set

Initialize the calculator The initialization will perform a global fit of the model to the data and build an Asimov data set.

It will then also fit the model to the Asimov data set to find the likelihood value of the Asimov data set nominalAsimov is an option for using Asimov data set obtained using nominal nuisance parameter values By default the nuisance parameters are fitted to the data NOTE: If a fit has been done before, one for speeding up could set all the initial parameters to the fit value and in addition set the null snapshot to the best fit

Definition at line 131 of file AsymptoticCalculator.cxx.

◆ IsA()

TClass * RooStats::AsymptoticCalculator::IsA ( ) const
inlineoverridevirtual
Returns
TClass describing current object

Reimplemented from RooStats::HypoTestCalculator.

Definition at line 125 of file AsymptoticCalculator.h.

◆ IsOneSidedDiscovery()

bool RooStats::AsymptoticCalculator::IsOneSidedDiscovery ( ) const
inline

Definition at line 86 of file AsymptoticCalculator.h.

◆ IsTwoSided()

bool RooStats::AsymptoticCalculator::IsTwoSided ( ) const
inline

Definition at line 85 of file AsymptoticCalculator.h.

◆ MakeAsimovData() [1/2]

RooAbsData * AsymptoticCalculator::MakeAsimovData ( const ModelConfig model,
const RooArgSet allParamValues,
RooArgSet asimovGlobObs 
)
static

Make a nominal Asimov data set from a model.

Parameters
modelModelConfig that contains the model pdf and the model parameters
allParamValuesThe parameters fo the model will be set to the values given in this set
[out]asimovGlobObsGlobal observables set to values satisfying the constraints
Returns
Asimov data set. The user takes ownership.

The parameter values (including the nuisance parameter) can result from a fit to data or be at the nominal values.

Definition at line 1323 of file AsymptoticCalculator.cxx.

◆ MakeAsimovData() [2/2]

RooAbsData * AsymptoticCalculator::MakeAsimovData ( RooAbsData realData,
const ModelConfig model,
const RooArgSet paramValues,
RooArgSet asimovGlobObs,
const RooArgSet genPoiValues = nullptr 
)
static

Make Asimov data.

Make the Asimov data from the ModelConfig and list of poi.

Parameters
realDataReal data
modelModel config defining the pdf and the parameters
paramValuesThe snapshot of POI and parameters used for finding the best nuisance parameter values (conditioned at these values)
[out]asimovGlobObsGlobal observables set to values satisfying the constraints
genPoiValuesOptional. A different set of POI values used for generating. By default the same POI are used for generating and for finding the nuisance parameters given an observed data set, a model and a snapshot of the poi.
Returns
The asimov data set. The user takes ownership.

Definition at line 1210 of file AsymptoticCalculator.cxx.

◆ SetAlternateModel()

void RooStats::AsymptoticCalculator::SetAlternateModel ( const ModelConfig model)
inlineoverridevirtual

Set the model for the alternate hypothesis.

Implements RooStats::HypoTestCalculator.

Definition at line 75 of file AsymptoticCalculator.h.

◆ SetData()

void RooStats::AsymptoticCalculator::SetData ( RooAbsData data)
inlineoverridevirtual

Set the DataSet.

Implements RooStats::HypoTestCalculator.

Definition at line 79 of file AsymptoticCalculator.h.

◆ SetNullModel()

void RooStats::AsymptoticCalculator::SetNullModel ( const ModelConfig nullModel)
inlineoverridevirtual

re-implementation of setters since they are needed to re-initialize the calculator

Implements RooStats::HypoTestCalculator.

Definition at line 71 of file AsymptoticCalculator.h.

◆ SetObsToExpected() [1/2]

bool AsymptoticCalculator::SetObsToExpected ( RooAbsPdf pdf,
const RooArgSet obs 
)
staticprotected

set observed value to the expected one works for Gaussian, Poisson or LogNormal assumes mean parameter value is the argument not constant and not depending on observables (if more than two arguments are not constant will use first one but print a warning !) need to iterate on the components of the Poisson to get n and nu (nu can be a RooAbsReal) (code from G.

Petrucciani and extended by L.M.)

Definition at line 954 of file AsymptoticCalculator.cxx.

◆ SetObsToExpected() [2/2]

bool AsymptoticCalculator::SetObsToExpected ( RooProdPdf prod,
const RooArgSet obs 
)
staticprotected

Inpspect a product pdf to find all the Poisson or Gaussian parts to set the observed values to expected ones.

Definition at line 915 of file AsymptoticCalculator.cxx.

◆ SetOneSided()

void RooStats::AsymptoticCalculator::SetOneSided ( bool  on)
inline

set test statistic for one sided (upper limits)

Definition at line 61 of file AsymptoticCalculator.h.

◆ SetOneSidedDiscovery()

void RooStats::AsymptoticCalculator::SetOneSidedDiscovery ( bool  on)
inline

set the test statistics for one-sided discovery

Definition at line 68 of file AsymptoticCalculator.h.

◆ SetPrintLevel()

void AsymptoticCalculator::SetPrintLevel ( int  level)
static

set print level (static function)

  • 0 minimal,
  • 1 normal,
  • 2 debug

Definition at line 86 of file AsymptoticCalculator.cxx.

◆ SetQTilde()

void RooStats::AsymptoticCalculator::SetQTilde ( bool  on)
inline

set using of qtilde, by default is controlled if RoORealVar is limited or not

Definition at line 90 of file AsymptoticCalculator.h.

◆ SetTwoSided()

void RooStats::AsymptoticCalculator::SetTwoSided ( )
inline

set the test statistics for two sided (in case of upper limits for discovery does not make really sense)

Definition at line 65 of file AsymptoticCalculator.h.

◆ Streamer()

void RooStats::AsymptoticCalculator::Streamer ( TBuffer )
overridevirtual

Reimplemented from RooStats::HypoTestCalculator.

◆ StreamerNVirtual()

void RooStats::AsymptoticCalculator::StreamerNVirtual ( TBuffer ClassDef_StreamerNVirtual_b)
inline

Definition at line 125 of file AsymptoticCalculator.h.

Member Data Documentation

◆ fAsimovData

RooAbsData* RooStats::AsymptoticCalculator::fAsimovData
mutableprivate

asimov data set

Definition at line 138 of file AsymptoticCalculator.h.

◆ fAsimovGlobObs

RooArgSet RooStats::AsymptoticCalculator::fAsimovGlobObs
mutableprivate

snapshot of Asimov global observables

Definition at line 139 of file AsymptoticCalculator.h.

◆ fBestFitParams

RooArgSet RooStats::AsymptoticCalculator::fBestFitParams
mutableprivate

snapshot of all best fitted Parameter values

Definition at line 141 of file AsymptoticCalculator.h.

◆ fBestFitPoi

RooArgSet RooStats::AsymptoticCalculator::fBestFitPoi
mutableprivate

snapshot of best fitted POI values

Definition at line 140 of file AsymptoticCalculator.h.

◆ fgPrintLevel

int AsymptoticCalculator::fgPrintLevel = 1
staticprivate

control print level (0 minimal, 1 normal, 2 debug)

Definition at line 134 of file AsymptoticCalculator.h.

◆ fIsInitialized

bool RooStats::AsymptoticCalculator::fIsInitialized
mutableprivate

! flag to check if calculator is initialized

Definition at line 132 of file AsymptoticCalculator.h.

◆ fNLLAsimov

double RooStats::AsymptoticCalculator::fNLLAsimov
mutableprivate

Definition at line 136 of file AsymptoticCalculator.h.

◆ fNLLObs

double RooStats::AsymptoticCalculator::fNLLObs
mutableprivate

Definition at line 135 of file AsymptoticCalculator.h.

◆ fNominalAsimov

bool RooStats::AsymptoticCalculator::fNominalAsimov
private

make Asimov at nominal parameter values

Definition at line 131 of file AsymptoticCalculator.h.

◆ fOneSided

bool RooStats::AsymptoticCalculator::fOneSided
private

for one sided PL test statistic (upper limits)

Definition at line 129 of file AsymptoticCalculator.h.

◆ fOneSidedDiscovery

bool RooStats::AsymptoticCalculator::fOneSidedDiscovery
mutableprivate

for one sided PL test statistic (for discovery)

Definition at line 130 of file AsymptoticCalculator.h.

◆ fUseQTilde

int RooStats::AsymptoticCalculator::fUseQTilde
mutableprivate

flag to indicate if using qtilde or not (-1 (default based on RooRealVar)), 0 false, 1 (true)

Definition at line 133 of file AsymptoticCalculator.h.

Libraries for RooStats::AsymptoticCalculator:

The documentation for this class was generated from the following files: