source: branches/peters_vector/lib/statistics/ROC.h @ 469

// $Id: ROC.h 469 2005-12-19 14:58:29Z peter $

#ifndef _theplu_statistics_roc_ 
#define _theplu_statistics_roc_ 

#include <c++_tools/classifier/Target.h>
#include <c++_tools/statistics/Score.h>

#include <utility>
#include <vector>

namespace theplu {
namespace statistics {  

  ///
  /// Class for ROC (Reciever Operating Characteristic). 
  ///   
  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 classifier::VectorAbstract& 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::VectorAbstract& value, 
                 const classifier::VectorAbstract& 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_; }  

    int target(const size_t i) const;
    inline size_t n(void) const { return vec_pair_.size(); }
    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<int, double> > vec_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