source:trunk/lib/statistics/Linear.h@430

Last change on this file since 430 was 430, checked in by Peter, 18 years ago

changed interface of Regression::Local

• Property svn:eol-style set to native
• Property svn:keywords set to Author Date Id Revision
File size: 2.5 KB
Line
1// $Id: Linear.h 430 2005-12-08 22:53:08Z peter$
2
3#ifndef _theplu_statistics_regression_linear_
4#define _theplu_statistics_regression_linear_
5
6#include <c++_tools/statistics/OneDimensional.h>
7#include <c++_tools/gslapi/vector.h>
8
9#include <cmath>
10
11namespace theplu {
12namespace statistics {
13namespace regression {
14
15  ///
16  /// @brief linear regression.
17  ///
18  /// @todo document
19  ///
20  class Linear : public OneDimensional
21  {
22
23  public:
24    ///
25    /// Default Constructor.
26    ///
27    inline Linear(void)
28      : OneDimensional(), alpha_(0), alpha_var_(0), beta_(0), beta_var_(0),
29        m_x_(0), s2_(0) {}
30
31    ///
32    /// Destructor
33    ///
34    inline virtual ~Linear(void) {};
35
36    ///
37    /// @return the parameter \f$\alpha \f$
38    ///
39    inline double alpha(void) const { return alpha_; }
40
41    ///
42    /// @return standard deviation of parameter \f$\alpha \f$
43    ///
44    inline double alpha_err(void) const { return sqrt(alpha_var_); }
45
46    ///
47    /// @return the parameter \f$\beta \f$
48    ///
49    inline double beta(void) const { return beta_; }
50
51    ///
52    /// @return standard deviation of parameter \f$\beta \f$
53    ///
54    inline double beta_err(void) const { return sqrt(beta_var_); }
55
56    ///
57    /// This function computes the best-fit linear regression
58    /// coefficients \f$(\alpha, \beta)\f$ of the model \f$y = 59 /// \alpha + \beta (x-m_x) \f$ from vectors \a x and \a y, by
60    /// minimizing \f$\sum{(y_i - \alpha - \beta (x-m_x))^2} \f$. By
61    /// construction \f$\alpha \f$ and \f$\beta \f$ are independent.
62    ///
63    void fit(const gslapi::vector& x, const gslapi::vector& y) ;
64
65    ///
66    /// Function predicting value using the linear model. \a y_err is
67    /// the expected deviation from the line for a new data point.
68    ///
69    void predict(const double x, double& y, double& y_err) const;
70
71    ///
72    /// @return prediction value and parameters
73    ///
74    std::ostream& print(std::ostream&) const;
75
76    ///
77    /// @return header for print()
78    ///
80
81    ///
82    /// Function returning the coefficient of determination,
83    /// i.e. fraction of variance explained by the linear model.
84    /// @todo implement r2's calculation in fit function
85    ///
86    inline double r2(void) const { return r2_; }
87
88  private:
89    ///
90    /// Copy Constructor. (not implemented)
91    ///
92    Linear(const Linear&);
93
94    double alpha_;
95    double alpha_var_;
96    double beta_;
97    double beta_var_;
98    double m_x_; // average of x values
99    double s2_; // var(y|x)
100    double r2_; // coefficient of determination
101  };
102
103}}} // of namespaces regression, statisitcs and thep
104
105#endif
Note: See TracBrowser for help on using the repository browser.