mcpdf.cpp

Go to the documentation of this file.

// $Id$
// Copyright (C) 2003 Klaas Gadeyne <first dot last at gmail dot com>
//
// This program is free software; you can redistribute it and/or modify
// it under the terms of the GNU Lesser General Public License as published by
// the Free Software Foundation; either version 2.1 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
// GNU Lesser General Public License for more details.
//
// You should have received a copy of the GNU Lesser General Public License
// along with this program; if not, write to the Free Software
// Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
//

// This file only contains template specialisation code
#include "mcpdf.h"

namespace BFL
{
  using namespace MatrixWrapper;


  // Template Specialisation for T =ColumnVector

  // Make sure no memory is allocated @runtime
  // Constructor
  template <> inline MCPdf<ColumnVector>::MCPdf(unsigned int num_samples, unsigned int dimension) :
      Pdf<ColumnVector>(dimension)
      , _CumSum(dimension)
      , _mean(dimension)
      , _diff(dimension)
      , _covariance(dimension)
      , _diffsum(dimension,dimension)
      {
        _SumWeights = 0;
        WeightedSample<ColumnVector> my_sample(dimension);
        _listOfSamples.insert(_listOfSamples.begin(),num_samples,my_sample);
        _CumPDF.insert(_CumPDF.begin(),num_samples+1,0.0);

       _los.assign(num_samples,WeightedSample<ColumnVector>(dimension));
       _it_los = _los.begin();
  #ifdef __CONSTRUCTOR__
        // if (num_samples > 0)
        cout << "MCPDF Constructor: NumSamples = " << _listOfSamples.size()
             << ", CumPDF Samples = " << _CumPDF.size()
             << ", _SumWeights = " << _SumWeights << endl;
  #endif // __CONSTRUCTOR__
      }

  // Make sure no memory is allocated @runtime
  // Copy constructor
  template <> inline
    MCPdf<ColumnVector>::MCPdf(const MCPdf & pdf) : Pdf<ColumnVector>(pdf)
    , _CumSum(pdf.DimensionGet())
    , _mean(pdf.DimensionGet())
    , _diff(pdf.DimensionGet())
    , _covariance(pdf.DimensionGet())
    , _diffsum(pdf.DimensionGet(),pdf.DimensionGet())
    {
      this->_listOfSamples = pdf._listOfSamples;
      this->_CumPDF = pdf._CumPDF;
      _SumWeights = pdf._SumWeights;
      this->_los = pdf._listOfSamples;
     _it_los = _los.begin();
#ifdef __CONSTRUCTOR__
      cout << "MCPDF Copy Constructor: NumSamples = " << _listOfSamples.size()
           << ", CumPDF Samples = " << _CumPDF.size()
           << ", SumWeights = " << _SumWeights << endl;
#endif // __CONSTRUCTOR__
    }

  template <> inline
  ColumnVector MCPdf<ColumnVector>::ExpectedValueGet (  ) const
  {
    _CumSum = 0.0;
    _los = _listOfSamples;
    for ( _it_los = _los.begin() ; _it_los != _los.end() ; _it_los++ )
         _CumSum += ( _it_los->ValueGet() * _it_los->WeightGet() );
    return _CumSum/_SumWeights;
  }


  template <> inline
  SymmetricMatrix MCPdf<ColumnVector>::CovarianceGet (  ) const
  {
    _mean = (this->ExpectedValueGet());
    _los = _listOfSamples;
    _diffsum = 0.0;
    // Actual calculation
    for (_it_los = _los.begin(); _it_los != _los.end(); _it_los++)
      {
        _diff = ( _it_los->ValueGet() - _mean);
        _diffsum += _diff * (_diff.transpose() * _it_los->WeightGet());
      }
    // Biased estimator!! (unbiased possible with weighted samples??)
    (_diffsum/_SumWeights).convertToSymmetricMatrix(_covariance);
    return _covariance;
  }






  // Template Specialisation for T =unsigned int
  template <> inline
  unsigned int MCPdf<unsigned int>::ExpectedValueGet (  ) const
  {
    unsigned int result;
    double CumSum = 0;
    _los = _listOfSamples;
    double current_weight;
    for ( _it_los = _los.begin() ; _it_los != _los.end() ; _it_los++ )
      {
        current_weight = _it_los->WeightGet();
        CumSum += ( ((double)_it_los->ValueGet()) * current_weight );
      }
    result = (unsigned int)((CumSum/_SumWeights) + 0.5);
    return result;
  }




  template <> inline
  SymmetricMatrix MCPdf<unsigned int>::CovarianceGet (  ) const
  {
    unsigned int mean = this->ExpectedValueGet();
    unsigned int diff;
    double diffsum, current_weight;
    _los = _listOfSamples;
    diffsum = 0.0;
    // Actual calculation
    for (_it_los = _los.begin(); _it_los != _los.end(); _it_los++)
      {
        current_weight = _it_los->WeightGet();
        diff = (_it_los->ValueGet() - mean);
        diffsum += (((double)(diff * diff)) * current_weight);
      }

    // Biased estimator!! (unbiased possible with weighted samples??)
    _covariance(1,1) = (diffsum / _SumWeights);
    return _covariance;
  }

}