Docs | All | Real | Category | PDF | DataSet | Plot | Container | Misc | Aux | User |
RooFitResultclass description - source file - inheritance tree (.pdf) class RooFitResult : public TNamed, public RooPrintable, public RooDirItem
protected: Data Membersprotected: Class DescriptionRooFitResult is a container class to hold the input and output of a PDF fit to a dataset. It contains: - Values of all constant parameters - Initial and final values of floating parameters with error - Correlation matrix and global correlation coefficients - NLL and EDM at mininum No references to the fitted PDF and dataset are stored RooFitResult(const char* name, const char* title) : TNamed(name,title), _constPars(0), _initPars(0), _finalPars(0), _globalCorr(0), _randomPars(0), _Lt(0) Constructor RooFitResult(const RooFitResult& other) : TNamed(other), RooPrintable(other), RooDirItem(), _status(other._status), _covQual(other._covQual), _numBadNLL(other._numBadNLL), _minNLL(other._minNLL), _edm(other._edm), _randomPars(0), _Lt(0) Copy constructor ~RooFitResult() Destructor void setConstParList(const RooArgList& list) Fill the list of constant parameters void setInitParList(const RooArgList& list) Fill the list of initial values of the floating parameters void setFinalParList(const RooArgList& list) Fill the list of final values of the floating parameters RooPlot* plotOn(RooPlot *frame, const char *parName1, const char *parName2, const char *options) const Add objects to a 2D plot that represent the fit results for the two named parameters. The input frame with the objects added is returned, or zero in case of an error. Which objects are added are determined by the options string which should be a concatenation of the following (not case sensitive): M - a marker at the best fit result E - an error ellipse calculated at 1-sigma using the error matrix at the minimum 1 - the 1-sigma error bar for parameter 1 2 - the 1-sigma error bar for parameter 2 B - the bounding box for the error ellipse H - a line and horizontal axis for reading off the correlation coefficient V - a line and vertical axis for reading off the correlation coefficient A - draw axes for reading off the correlation coefficients with the H or V options You can change the attributes of objects in the returned RooPlot using the various RooPlot::getAttXxx(name) member functions, e.g. plot->getAttLine("contour")->SetLineStyle(kDashed); Use plot->Print() for a list of all objects and their names (unfortunately most of the ROOT builtin graphics objects like TLine are unnamed). Drag the left mouse button along the labels of either axis button to interactively zoom in a plot. const RooArgList& randomizePars() const Return a list of floating parameter values that are perturbed from the final fit values by random amounts sampled from the covariance matrix. The returned object is overwritten with each call and belongs to the RooFitResult. Uses the "square root method" to decompose the covariance matrix, which makes inverting it unnecessary. Double_t correlation(const char* parname1, const char* parname2) const Return the correlation between parameters 'par1' and 'par2' const RooArgList* correlation(const char* parname) const Return the set of correlation coefficients of parameter 'par' with all other floating parameters Double_t globalCorr(const char* parname) Return the global correlation of the named parameter const RooArgList* globalCorr() Return the list of all global correlations Double_t correlation(Int_t row, Int_t col) const Return a correlation matrix element addressed with numeric indices. Double_t covariance(Int_t row, Int_t col) const Return the covariance matrix element addressed with numeric indices. void printToStream(ostream& os, PrintOption opt, TString indent) const Print fit result to stream 'os'. In Verbose mode, the contant parameters and the initial and final values of the floating parameters are printed. Standard mode only the final values of the floating parameters are printed void fillCorrMatrix() Extract the correlation matrix and the global correlation coefficients from the MINUIT memory buffer and fill the internal arrays. RooFitResult* lastMinuitFit(const RooArgList& varList) Verify length of supplied varList Inline FunctionsTObject* clone() const void Print(Option_t* options = "0") const Int_t status() const Int_t covQual() const Int_t numInvalidNLL() const Double_t edm() const Double_t minNll() const const RooArgList& constPars() const const RooArgList& floatParsInit() const const RooArgList& floatParsFinal() const const RooArgList* correlation(const char* parname) const const RooArgList* globalCorr() const RooPlot* plotOn(RooPlot* plot, const char* parName1, const char* parName2, const char* options = "ME") const void setMinNLL(Double_t val) void setEDM(Double_t val) void setStatus(Int_t val) void setCovQual(Int_t val) void setNumInvalidNLL(Int_t val) Double_t correlation(Int_t row, Int_t col) const TClass* Class() TClass* IsA() const void ShowMembers(TMemberInspector& insp, char* parent) void Streamer(TBuffer& b) void StreamerNVirtual(TBuffer& b)
|
Page maintained by Wouter Verkerke and David Kirkby |