ROOT logo
ROOT » ROOFIT » ROOSTATS » RooStats::NeymanConstruction

class RooStats::NeymanConstruction: public RooStats::IntervalCalculator

NeymanConstruction is a concrete implementation of the NeymanConstruction interface that, as the name suggests, performs a NeymanConstruction. It produces a RooStats::PointSetInterval, which is a concrete implementation of the ConfInterval interface.

The Neyman Construction is not a uniquely defined statistical technique, it requires that one specify an ordering rule or ordering principle, which is usually incoded by choosing a specific test statistic and limits of integration (corresponding to upper/lower/central limits). As a result, this class must be configured with the corresponding information before it can produce an interval. Common configurations, such as the Feldman-Cousins approach, can be enforced by other light weight classes.

The Neyman Construction considers every point in the parameter space independently, no assumptions are made that the interval is connected or of a particular shape. As a result, the PointSetInterval class is used to represent the result. The user indicate which points in the parameter space to perform the constrution by providing a PointSetInterval instance with the desired points.

This class is fairly light weight, because the choice of parameter points to be considered is factorized and so is the creation of the sampling distribution of the test statistic (which is done by a concrete class implementing the DistributionCreator interface). As a result, this class basically just drives the construction by:

  • using a DistributionCreator to create the SamplingDistribution of a user-defined test statistic for each parameter point of interest,
  • defining the acceptance region in the data by finding the thresholds on the test statistic such that the integral of the sampling distribution is of the appropriate size and consistent with the limits of integration (eg. upper/lower/central limits),
  • and finally updating the PointSetInterval based on whether the value of the test statistic evaluated on the data are in the acceptance region.

Function Members (Methods)

static TClass*Class()
static TClass*RooStats::IntervalCalculator::Class()
virtual Double_tConfidenceLevel() const
virtual Double_tRooStats::IntervalCalculator::ConfidenceLevel() const
voidCreateConfBelt(bool flag = true)
virtual TList*GenSamplingDistribution(const char* asciiFilePat = 0) const
virtual RooStats::ConfInterval*GetInterval() const
virtual RooStats::ConfInterval*RooStats::IntervalCalculator::GetInterval() const
virtual RooStats::ConfInterval*GetInterval(const char* asciiFilePat) const
virtual RooStats::ConfInterval*GetIntervalUsingList() const
virtual TClass*IsA() const
virtual TClass*RooStats::IntervalCalculator::IsA() const
RooStats::NeymanConstructionNeymanConstruction(const RooStats::NeymanConstruction&)
RooStats::IntervalCalculator&RooStats::IntervalCalculator::operator=(const RooStats::IntervalCalculator&)
virtual RooStats::ConfInterval*Run(TList* SamplingList) const
voidSaveBeltToFile(bool flag = true)
virtual voidSetConfidenceLevel(Double_t cl)
virtual voidRooStats::IntervalCalculator::SetConfidenceLevel(Double_t cl)
virtual voidSetData(RooAbsData& data)
virtual voidSetData(const char* name)
virtual voidRooStats::IntervalCalculator::SetData(RooAbsData&)
virtual voidRooStats::IntervalCalculator::SetData(const char* name)
voidSetLeftSideTailFraction(Double_t leftSideFraction = 0.)
virtual voidSetNuisanceParameters(RooArgSet& set)
virtual voidRooStats::IntervalCalculator::SetNuisanceParameters(RooArgSet&)
voidSetParameterPointsToTest(RooAbsData& pointsToTest)
virtual voidSetParameters(RooArgSet& set)
virtual voidRooStats::IntervalCalculator::SetParameters(RooArgSet&)
virtual voidSetPdf(RooAbsPdf& pdf)
virtual voidSetPdf(const char* name)
virtual voidRooStats::IntervalCalculator::SetPdf(RooAbsPdf&)
virtual voidRooStats::IntervalCalculator::SetPdf(const char* name)
virtual voidSetTestSize(Double_t size)
virtual voidRooStats::IntervalCalculator::SetTestSize(Double_t size)
voidSetTestStatSampler(RooStats::TestStatSampler& distCreator)
virtual voidSetWorkspace(RooWorkspace& ws)
virtual voidRooStats::IntervalCalculator::SetWorkspace(RooWorkspace& ws)
virtual voidShowMembers(TMemberInspector& insp, char* parent)
virtual voidRooStats::IntervalCalculator::ShowMembers(TMemberInspector& insp, char* parent)
virtual Double_tSize() const
virtual Double_tRooStats::IntervalCalculator::Size() const
virtual voidStreamer(TBuffer& b)
virtual voidRooStats::IntervalCalculator::Streamer(TBuffer& b)
voidStreamerNVirtual(TBuffer& b)
voidRooStats::IntervalCalculator::StreamerNVirtual(TBuffer& b)
voidUseAdaptiveSampling(bool flag = true)

Data Members

boolfAdaptiveSamplingcontrols use of adaptive sampling algorithm
boolfCreateBeltcontrols use if ConfidenceBelt should be saved to a TFile
const char*fDataNamename of data set in workspace
RooArgSet*fNuisParamsRooArgSet specifying nuisance parameters for interval
Bool_tfOwnsWorkspaceflag if this object owns its workspace
RooArgSet*fPOIRooArgSet specifying parameters of interest for interval
const char*fPdfNamename of common PDF in workspace
boolfSaveBeltToFilecontrols use if ConfidenceBelt should be saved to a TFile
Double_tfSizesize of the test (eg. specified rate of Type I error)
RooWorkspace*fWSa workspace that owns all the components to be used by the calculator

Class Charts

Inheritance Inherited Members Includes Libraries
Class Charts

Function documentation

 default constructor
 default constructor
  if(fOwnsWorkspace && fWS) delete fWS;
  if(fConfBelt) delete fConfBelt;
ConfInterval* GetInterval() const
 Main interface to get a RooStats::ConfInterval.
 It constructs a RooStats::SetInterval.
TList* GenSamplingDistribution(const char* asciiFilePat = 0) const
This method generates the sampling distribution for each point of the study. If a file path
is provided, the distribution is saved in a root file. Returns the list of the distributions
for each point.
ConfInterval* GetIntervalUsingList() const
 Main interface to get a RooStats::ConfInterval.
 It constructs a RooStats::PointSetInterval.
ConfInterval* GetInterval(const char* asciiFilePat) const
This method returns a confidence interval exactly like GetInterval(), but
instead of generating the sampling disribution (long computation) it takes
the distribution from the file provided
ConfInterval* Run(TList* SamplingList) const
Main method to perform the interval calculation
void SetTestStatSampler(RooStats::TestStatSampler& distCreator)
 in addition to interface we also need:
 Set the TestStatSampler (eg. ToyMC or FFT, includes choice of TestStatistic)
{fTestStatSampler = &distCreator;}
void SetLeftSideTailFraction(Double_t leftSideFraction = 0.)
{fLeftSideFraction = leftSideFraction;}
void SetParameterPointsToTest(RooAbsData& pointsToTest)
 User-defined set of points to test
Double_t Size() const
 This class can make regularly spaced scans based on range stored in RooRealVars.
 Choose number of steps for a rastor scan (common for each dimension)
      void SetNumSteps(Int_t);
 This class can make regularly spaced scans based on range stored in RooRealVars.
 Choose number of steps for a rastor scan (specific for each dimension)
      void SetNumSteps(map<RooAbsArg, Int_t>)
 Get the size of the test (eg. rate of Type I error)
{return fSize;}
Double_t ConfidenceLevel() const
 set a workspace that owns all the necessary components for the analysis
{return 1.-fSize;}
void SetWorkspace(RooWorkspace& ws)
{fWS = &ws;}
void SetData(RooAbsData& data)
 Set the DataSet, add to the the workspace if not already there
void SetPdf(RooAbsPdf& pdf)
 Set the Pdf, add to the the workspace if not already there
void SetParameters(RooArgSet& set)
 specify the parameters of interest in the interval
{fPOI = &set;}
void SetNuisanceParameters(RooArgSet& set)
 set the size of the test (rate of Type I error) ( Eg. 0.05 for a 95% Confidence Interval)
{fNuisParams = &set;}
void SetTestSize(Double_t size)
 set the confidence level for the interval (eg. 0.95 for a 95% Confidence Interval)
{fSize = size;}
void SetConfidenceLevel(Double_t cl)
{fSize = 1.-cl;}
ConfidenceBelt* GetConfidenceBelt()
{return fConfBelt;}
void UseAdaptiveSampling(bool flag = true)
void SaveBeltToFile(bool flag = true)
void CreateConfBelt(bool flag = true)
{fCreateBelt = flag;}