source: trunk/yat/statistics/AveragerPairWeighted.h @ 1275

Last change on this file since 1275 was 1275, checked in by Jari Häkkinen, 13 years ago

Updating copyright statements.

  • Property svn:eol-style set to native
  • Property svn:keywords set to Id
File size: 4.7 KB
Line 
1#ifndef _theplu_yat_statistics_averagerpairweighted_
2#define _theplu_yat_statistics_averagerpairweighted_
3
4// $Id: AveragerPairWeighted.h 1275 2008-04-11 06:10:12Z jari $
5
6/*
7  Copyright (C) 2005 Peter Johansson, Markus Ringnér
8  Copyright (C) 2006, 2007 Jari Häkkinen, Peter Johansson, Markus Ringnér
9  Copyright (C) 2008 Peter Johansson
10
11  This file is part of the yat library, http://trac.thep.lu.se/yat
12
13  The yat library is free software; you can redistribute it and/or
14  modify it under the terms of the GNU General Public License as
15  published by the Free Software Foundation; either version 2 of the
16  License, or (at your option) any later version.
17
18  The yat library is distributed in the hope that it will be useful,
19  but WITHOUT ANY WARRANTY; without even the implied warranty of
20  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
21  General Public License for more details.
22
23  You should have received a copy of the GNU General Public License
24  along with this program; if not, write to the Free Software
25  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
26  02111-1307, USA.
27*/
28
29#include "AveragerWeighted.h"
30
31#include "yat/utility/iterator_traits.h"
32#include "yat/utility/yat_assert.h"
33
34#include <cmath>
35#include <stdexcept>
36
37namespace theplu{
38namespace yat{
39namespace statistics{
40  ///
41  /// @brief Class for taking care of mean and covariance of two variables in
42  /// a weighted manner.
43  ///
44  /// \see \ref weighted_statistics
45  ///
46  /// If nothing else stated, each function fulfills the
47  /// following:<br> <ul><li>Setting a weight to zero corresponds to
48  /// removing the data point from the dataset.</li><li> Setting all
49  /// weights to unity, the yields the same result as from
50  /// corresponding function in AveragerPair.</li><li> Rescaling weights
51  /// does not change the performance of the object.</li></ul>
52  ///
53  /// @see Averager AveragerWeighted AveragerPair
54  ///
55  class AveragerPairWeighted
56  {
57  public:
58
59    ///
60    /// @brief The default constructor
61    ///
62    AveragerPairWeighted(void);
63
64    ///
65    /// Adding a pair of data points with value \a x and \a y, and
66    /// their weights. If either of the weights are zero the addition
67    /// is ignored
68    ///
69    void  add(const double x, const double y, 
70              const double wx, const double wy);
71
72    ///
73    /// @brief %Pearson correlation coefficient.
74    ///
75    /// @return \f$ \frac{\sum w_xw_y (x-m_x)(y-m_y)}{\sqrt{\sum
76    /// w_xw_y (y-m_y)^2\sum w_xw_y (y-m_y)^2}} \f$ where m is
77    /// calculated as \f$ m_x = \frac {\sum w_xw_yx}{\sum w} \f$
78    ///
79    double correlation(void) const;
80 
81    ///
82    /// \f$ \frac{\sum w_xw_y (x-m_x)(y-m_y)}{\sum w_xw_y} \f$ where m
83    /// is calculated as \f$ m_x = \frac {\sum w_xw_yx}{\sum w} \f$
84    ///
85    double covariance(void) const;
86
87    /**
88       @return \f$ \frac{\sum w_xw_y(x-y)^2}{\sum w_xw_y} \f$
89    */
90    double msd(void) const;
91
92    /**
93       @return \f$ \frac{\left(\sum w_x w_y\right)^2}{\sum w_x^2w_y^2} \f$
94    */ 
95    double n(void) const;
96
97    ///
98    /// @brief Reset everything to zero
99    ///
100    void reset(void);
101
102    ///
103    /// @return \f$ \sum w_xw_y \f$
104    ///
105    double sum_w(void) const;
106
107    ///
108    /// @return \f$ \sum w_xw_yxy \f$
109    ///
110    double sum_xy(void) const;
111
112    ///
113    /// @return \f$ \sum w_xw_y (x-m_x)(y-m_y) \f$ where m is calculated as
114    /// \f$ m_x = \frac {\sum w_xw_yx}{\sum w} \f$
115    ///
116    double sum_xy_centered(void) const;
117
118    ///
119    /// @note the weights are calculated as \f$ w = w_x * w_y \f$
120    ///
121    /// @return AveragerWeighted for x
122    ///
123    const AveragerWeighted& x_averager(void) const;
124
125    ///
126    /// @note the weights are calculated as \f$ w = w_x * w_y \f$
127    ///
128    /// @return AveragerWeighted for y
129    ///
130    const AveragerWeighted& y_averager(void) const;
131
132  private:
133    AveragerWeighted x_; // weighted averager with w = w_x*w_y
134    AveragerWeighted y_; // weighted averager with w = w_x*w_y
135    double wxy_;
136    double w_;
137
138  };
139
140  /**
141     \brief adding a ranges of values to AveragerPairWeighted \a ap
142   */
143  template <class Iter1, class Iter2>
144  void add(AveragerPairWeighted& ap, Iter1 first1, Iter1 last1, Iter2 first2)
145  {
146    for ( ; first1 != last1; ++first1, ++first2) {
147      ap.add(utility::iterator_traits<Iter1>().data(first1), 
148             utility::iterator_traits<Iter2>().data(first2),
149             utility::iterator_traits<Iter1>().weight(first1), 
150             utility::iterator_traits<Iter2>().weight(first2));
151    }
152  }
153
154
155  /**
156     \brief adding four ranges of values to AveragerPairWeighted \a ap
157   */
158  template <class Iter1, class Iter2, class Iter3, class Iter4>
159  void add(AveragerPairWeighted& ap, Iter1 x, Iter1 xlast, Iter2 y, Iter3 wx, 
160           Iter4 wy)
161  {
162    utility::check_iterator_is_unweighted(x);
163    utility::check_iterator_is_unweighted(y);
164    utility::check_iterator_is_unweighted(wx);
165    utility::check_iterator_is_unweighted(wy);
166    while (x!=xlast) {
167      ap.add(*x, *y, *wx, *wy);
168      ++x;
169      ++y;
170      ++wx;
171      ++wy;
172    }
173  }
174
175
176}}} // of namespace statistics, yat, and theplu
177
178#endif
Note: See TracBrowser for help on using the repository browser.