LauExponentialPdf.cc

Go to the documentation of this file.

// Copyright University of Warwick 2011 - 2013.
// Distributed under the Boost Software License, Version 1.0.
// (See accompanying file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)

// Authors:
// Thomas Latham
// John Back
// Paul Harrison

#include <iostream>
#include <vector>
using std::cout;
using std::cerr;
using std::endl;
using std::vector;

#include "TMath.h"
#include "TSystem.h"

#include "LauConstants.hh"
#include "LauExponentialPdf.hh"

ClassImp(LauExponentialPdf)


LauExponentialPdf::LauExponentialPdf(const TString& theVarName, const vector<LauAbsRValue*>& params, Double_t minAbscissa, Double_t maxAbscissa) :
        LauAbsPdf(theVarName, params, minAbscissa, maxAbscissa),
        slope_(0)
{
        // Constructor for the Exponential PDF.
        //
        // The parameter in param is the slope of the exponential ie exp(slope*x).
        // The last two arguments specify the range in which the PDF is defined, and the PDF
        // will be normalised w.r.t. these limits.

        slope_ = this->findParameter("slope");

        if ((this->nParameters() != 1) || (slope_ == 0)) {
                cerr<<"ERROR in LauExponentialPdf constructor: LauExponentialPdf requires 1 parameter: \"slope\"."<<endl;
                gSystem->Exit(EXIT_FAILURE);
        }

        // Cache the normalisation factor
        this->calcNorm();
}

LauExponentialPdf::~LauExponentialPdf() 
{
        // Destructor
}

void LauExponentialPdf::calcLikelihoodInfo(const LauAbscissas& abscissas)
{
        // Check that the given abscissa is within the allowed range
        if (!this->checkRange(abscissas)) {
                gSystem->Exit(EXIT_FAILURE);
        }

        // Get our abscissa
        Double_t abscissa = abscissas[0];

        // Get the up to date parameter values
        Double_t slope = slope_->unblindValue();

        // Calculate the value of the Exponential for the given value of the abscissa.

        Double_t exponent(0.0);
        if(slope != 0){
                exponent = slope*abscissa;
        }

        Double_t value = TMath::Exp(exponent);
        this->setUnNormPDFVal(value);

        // if the parameters are floating then we
        // need to recalculate the normalisation
        if (!this->cachePDF() && !this->withinNormCalc() && !this->withinGeneration()) {
                this->calcNorm();
        }

}

void LauExponentialPdf::calcNorm() 
{
        // Get the up to date parameter values
        Double_t slope = slope_->unblindValue();

        // Calculate the normalisation of the exponential and cache it.
        Double_t norm(0.0);

        if(slope != 0){
                norm = (1/slope)*(TMath::Exp(slope*this->getMaxAbscissa()) - TMath::Exp(slope*this->getMinAbscissa())) ;
        }

        this->setNorm(norm);
}

void LauExponentialPdf::calcPDFHeight(const LauKinematics* /*kinematics*/)
{
        if (this->heightUpToDate()) {
                return;
        }

        // Get the up to date parameter values
        Double_t slope = slope_->unblindValue();

        LauAbscissas maxPoint(1);
        maxPoint[0] = 0;

        // Calculate the PDF height for the Exponential function.
        if (slope > 0) {
                maxPoint[0] = this->getMaxAbscissa();
        } else if (slope < 0) {
                maxPoint[0] = this->getMinAbscissa();
        }
        this->calcLikelihoodInfo(maxPoint);

        Double_t height = this->getUnNormLikelihood();
        this->setMaxHeight(height);
}