source: trunk/c++_tools/statistics/ROC.h @ 675

#ifndef _theplu_statistics_roc_ 
#define _theplu_statistics_roc_ 

// $Id: ROC.h 675 2006-10-10 12:08:45Z jari $

/*
  Copyright (C) The authors contributing to this file.

  This file is part of the yat library, http://lev.thep.lu.se/trac/yat

  The yat library is free software; you can redistribute it and/or
  modify it under the terms of the GNU General Public License as
  published by the Free Software Foundation; either version 2 of the
  License, or (at your option) any later version.

  The yat library 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
  General Public License for more details.

  You should have received a copy of the GNU 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.
*/

#include "yat/classifier/Target.h"
#include "yat/statistics/Score.h"

#include <utility>
#include <vector>

namespace theplu {
  namespace utility {
    class vector;
  }
namespace statistics {  

  ///
  /// Class for ROC (Reciever Operating Characteristic). 
  ///   
  /// As the area under an ROC curve is equivalent to Mann-Whitney U
  /// statistica, this class can be used to perform a Mann-Whitney
  /// U-test (aka Wilcoxon).
  ///
  class ROC : public Score
  {
  
  public:
    ///
    /// Default constructor
    ///
    ROC(bool absolute=true);
         
    ///
    /// Destructor 
    ///
    virtual ~ROC(void) {};
         
    /// Function taking \a value, \a target (+1 or -1) and vector
    /// defining what samples to use. The score is equivalent to
    /// Mann-Whitney statistics.
    /// @return the area under the ROC curve. If the area is less
    /// than 0.5 and absolute=true, 1-area is returned. Complexity is
    /// \f$ N\log N \f$ where \f$ N \f$ is number of samples.
    ///
    double score(const classifier::Target& target, 
                 const utility::vector& value); 
    
    /** 
        Function taking values, target, weight and a vector defining
        what samples to use. The area is defines as \f$ \frac{\sum
        w^+w^-}{\sum w^+w^-}\f$, where the sum in the numerator goes
        over all pairs where value+ is larger than value-. The
        denominator goes over all pairs. If target is equal to 1,
        sample belonges to class + otherwise sample belongs to class
        -. @return wheighted version of area under the ROC curve. If
        the area is less than 0.5 and absolute=true, 1-area is
        returned. Complexity is \f$ N^2 \f$ where \f$ N \f$ is number
        of samples.
    */
    double score(const classifier::Target& target, 
                 const classifier::DataLookupWeighted1D& value); 
       

    /** 
        Function taking values, target, weight and a vector defining
        what samples to use. The area is defines as \f$ \frac{\sum
        w^+w^-}{\sum w^+w^-}\f$, where the sum in the numerator goes
        over all pairs where value+ is larger than value-. The
        denominator goes over all pairs. If target is equal to 1,
        sample belonges to class + otherwise sample belongs to class
        -. @return wheighted version of area under the ROC curve. If
        the area is less than 0.5 and absolute=true, 1-area is
        returned. Complexity is \f$ N^2 \f$ where \f$ N \f$ is number
        of samples.
    */
    double score(const classifier::Target& target, 
                 const utility::vector& value, 
                 const utility::vector& weight); 
       

    ///
    ///Calculates the p-value, i.e. the probability of observing an
    ///area equally or larger if the null hypothesis is true. If P is
    ///near zero, this casts doubt on this hypothesis. The null
    ///hypothesis is that the values from the 2 classes are generated
    ///from 2 identical distributions. The alternative is that the
    ///median of the first distribution is shifted from the median of
    ///the second distribution by a non-zero amount. If the smallest
    ///group size is larger than minimum_size (default = 10), then P
    ///is calculated using a normal approximation.  @return the
    ///one-sided p-value( if absolute true is used this is equivalent
    ///to the two-sided p-value.)
    ///
    double p_value(void) const;
    
    ///
    /// minimum_size is the threshold for when a normal
    /// approximation is used for the p-value calculation.
    ///
    /// @return reference to minimum_size
    ///
    inline u_int& minimum_size(void){ return minimum_size_; }  

    ///
    /// Function returning true if target is positive (binary()) for
    /// the sample with ith lowest data value, so i=0 corresponds to
    /// the sample with the lowest data value and i=n()-1 the sample
    /// with highest data value.
    ///
    bool target(const size_t i) const;

    ///
    /// @return number of samples
    ///
    inline size_t n(void) const { return vec_pair_.size(); }

    ///
    /// @return number of positive samples (Target.binary()==true)
    ///
    inline size_t n_pos(void) const { return nof_pos_; }

  private:
    
    /// Implemented as in MatLab 13.1
    double get_p_approx(const double) const;

    /// Implemented as in MatLab 13.1
    double get_p_exact(const double, const double, const double) const;

    double area_;
    u_int minimum_size_;
    u_int nof_pos_;
    std::vector<std::pair<bool, double> > vec_pair_; // class-value-pair
  };

  ///
  /// The output operator for the ROC class. The output is an Nx2
  /// matrix, where the first column is the sensitivity and second
  /// is the specificity.
  ///
  std::ostream& operator<< (std::ostream& s, const ROC&);


}} // of namespace statistics and namespace theplu

#endif