source: trunk/yat/statistics/Histogram.h @ 1701

Last change on this file since 1701 was 1487, checked in by Jari Häkkinen, 13 years ago

Addresses #436. GPL license copy reference should also be updated.

  • Property svn:eol-style set to native
  • Property svn:keywords set to Author Date Id Revision
File size: 6.2 KB
Line 
1#ifndef _theplu_yat_statistics_histogram_
2#define _theplu_yat_statistics_histogram_
3
4// $Id: Histogram.h 1487 2008-09-10 08:41:36Z jari $
5
6/*
7  Copyright (C) 2004 Jari Häkkinen
8  Copyright (C) 2005 Jari Häkkinen, Peter Johansson
9  Copyright (C) 2006 Jari Häkkinen
10  Copyright (C) 2007 Jari Häkkinen, Peter Johansson
11  Copyright (C) 2008 Peter Johansson
12
13  This file is part of the yat library, http://dev.thep.lu.se/yat
14
15  The yat library is free software; you can redistribute it and/or
16  modify it under the terms of the GNU General Public License as
17  published by the Free Software Foundation; either version 3 of the
18  License, or (at your option) any later version.
19
20  The yat library is distributed in the hope that it will be useful,
21  but WITHOUT ANY WARRANTY; without even the implied warranty of
22  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
23  General Public License for more details.
24
25  You should have received a copy of the GNU General Public License
26  along with yat. If not, see <http://www.gnu.org/licenses/>.
27*/
28
29#include "AveragerWeighted.h"
30#include "yat/utility/iterator_traits.h"
31
32#include <string>
33#include <vector>
34
35namespace theplu {
36namespace yat {
37namespace statistics {
38
39  ///
40  /// @brief Histograms provide a convenient way of presenting the
41  /// distribution of a set of data.
42  ///
43  /// A histogram consists of a set of
44  /// bins which count the number of events falling into these
45  /// bins. Currently only one dimensional histograms with uniformly
46  /// spaced bins are supported.
47  ///
48  class Histogram
49  {
50  public:
51
52    ///
53    /// The default constructor.
54    ///
55    Histogram(void);
56
57    ///
58    /// The copy constructor.
59    ///
60    Histogram(const Histogram&);
61
62    ///
63    /// Construct a histogram object that covers \f$(xmin,xmax]\f$
64    /// with the bin spacing \f$(xmax-xmin)/n\f$.
65    ///
66    Histogram(const double xmin, const double xmax, const size_t n);
67
68    virtual ~Histogram(void);
69
70    ///
71    /// Update the histogram by adding \a weight to the bin whose
72    /// range contains the observation \a x. No bins are updated when
73    /// \a x lies outside the range of the histogram but the value is
74    /// added to the overall integral of the histogram.
75    ///
76    /// @short Add a data point to the histogram.
77    ///
78    /// @return 0 if \a x lies within the range of the histogram, -1
79    /// if \a x is smaller than the lower limit of the histogram, and
80    /// similarly, 1 is returned if \a x is greater than or equal to
81    /// the upper limit.
82    ///
83    int add(const double x,const double weight=1.0);
84
85    ///
86    /// Gives access to the AveragerWeighted object that keeps track of
87    /// average of all events presented to the histogram.
88    ///
89    /// @short Average of all events presented to the histogram.
90    ///
91    /// @return A const reference to an AveragerWeighted object.
92    ///
93    const AveragerWeighted& averager_all(void) const;
94
95    ///
96    /// Gives access to the AveragerWeighted object that keeps track of
97    /// average of events that fits within the histogram lower and
98    /// upper limits. This function is equivalent to averager().
99    ///
100    /// @short Average of events fitting within histogram.
101    ///
102    /// @return A const reference to an AveragerWeighted object.
103    ///
104    const AveragerWeighted& averager_histogram(void) const;
105
106    ///
107    /// @return The number of bins in the histogram
108    ///
109    size_t nof_bins(void) const;
110
111    ///
112    ///  There are two ways to normalize the counts.
113    ///
114    /// If choice is true: The normalized count is the count in a
115    /// bin divided by the total number of observations. In this
116    /// case the relative counts are normalized to sum to unity (
117    /// minus values outside histogram).  This is the intuitive case
118    /// where the height of the histogram bar represents the
119    /// proportion of the data in each class.
120    ///
121    /// If choice is false: The normalized count is the count in the
122    /// class divided by the number of observations times the bin
123    /// width. For this normalization, the area (or integral) under
124    /// the histogram is equal to unity (minus the missing area
125    /// corresponding to counts outside histogram). From a
126    /// probabilistic point of view, this normalization results in a
127    /// relative histogram that is most akin to the probability
128    /// density function If you want to overlay a probability density
129    /// on top of the histogram, use this normalization. Although this
130    /// normalization is less intuitive (relative frequencies greater
131    /// than 1 are quite permissible), it is the appropriate
132    /// normalization if you are using the histogram to model a
133    /// probability density function.
134    ///
135    /// @short Normalizing the histogram
136    ///
137    void normalize(bool choice = true);
138
139    ///
140    /// @return The value in the middle of bin \a k.
141    ///
142    /// @note No check is done that \a k is within the size of the
143    /// histogram.
144    ///
145    double observation_value(const size_t k) const;
146
147    ///
148    /// Set everyting to default values, here it means that everything
149    /// is set to zero except the boundary values that are kept.
150    ///
151    void reset(void);
152
153    ///
154    /// @return The width of the bins in the histogram.
155    ///
156    double spacing(void) const;
157
158    ///
159    /// @return The histogram upper boundary.
160    ///
161    /// @note The upper boundary value is outside the histogram.
162    ///
163    double xmax(void) const;
164
165    ///
166    /// @return The histogram lower boundary.
167    ///
168    /// @note The lower boundary value is inside the histogram.
169    ///
170    double xmin(void) const;
171
172    ///
173    /// @return The count of bin \a k in the histogram.
174    ///
175    double operator[](size_t k) const;
176
177    ///
178    /// The assignment operator
179    ///
180    const Histogram& operator=(const Histogram&);
181
182  private:
183    // Returns zero if outside boundaries
184    size_t bin(double d);
185
186    std::vector<double> histogram_;
187    double xmax_;
188    double xmin_;
189    statistics::AveragerWeighted sum_all_;      // average of all data
190    statistics::AveragerWeighted sum_histogram_;// average of data in histogram
191  };
192
193  /**
194     Add a range [first, last) of values to Histogram.
195   */
196  template<typename ForwardIterator>
197  void add(Histogram& h, 
198           ForwardIterator first, ForwardIterator last)
199  {
200    while (first!=last) {
201      h.add(utility::iterator_traits<ForwardIterator>().data(),
202            utility::iterator_traits<ForwardIterator>().weight());
203      ++first;
204    }
205  }
206
207///
208/// The Histogram output operator
209///
210std::ostream& operator<<(std::ostream& s,const Histogram&);
211
212}}} // of namespace statistics, yat, and theplu
213
214#endif
Note: See TracBrowser for help on using the repository browser.