# source:trunk/src/RegressionNaive.h@286

Last change on this file since 286 was 286, checked in by Peter, 18 years ago

added default as in base class

• Property svn:eol-style set to native
• Property svn:keywords set to Author Date Id Revision
File size: 2.3 KB
Line
1// $Id: RegressionNaive.h 286 2005-04-21 21:49:31Z peter$
2
3#ifndef _theplu_statistics_regression_naive_
4#define _theplu_statistics_regression_naive_
5
6// C++ tools include
7/////////////////////
8#include "Averager.h"
9#include "Regression.h"
10#include "vector.h"
11#include "WeightedAverager.h"
12// Standard C++ includes
13////////////////////////
14//#include <gsl/gsl_fit.h>
15#include <iostream>
16#include <utility>
17
18
19
20namespace theplu {
21namespace statistics {
22
23  ///
24  /// Class for Regression.
25  ///
26
27  class RegressionNaive : public Regression
28  {
29
30  public:
31    ///
32    /// Default Constructor.
33    ///
34    RegressionNaive(void);
35
36    ///
37    /// Copy Constructor. (not implemented)
38    ///
39    RegressionNaive(const RegressionNaive&);
40
41    ///
42    /// Destructor
43    ///
44    virtual ~RegressionNaive(void) {};
45
46    ///
47    /// This function computes the best-fit for the naive model \f$y 48 /// = m \f$ from vectors \a x and \a y, by minimizing \f$49 /// \sum{(y_i-m)^2} \f$. This function is the same as using the
50    /// weighted version with unity weights.
51    ///
52    void fit(const gslapi::vector& x, const gslapi::vector& y);
53
54    ///
55    /// This function computes the best-fit for the naive model \f$y 56 /// = m \f$ from vectors \a x and \a y, by minimizing \f$\sum 57 /// w_i(y_i-m)^2 \f$. The weight \f$w_i \f$ is proportional to
58    /// the inverse of the variance for \f$y_i \f$
59    ///
60    void fit(const gslapi::vector& x,
61             const gslapi::vector& y,
62             const gslapi::vector& w);
63
64    ///
65    /// Function predicting value using the naive model. \a y_err is
66    /// the expected deviation from the line for a new data point. The
67    /// weight for the new point can be specified. A smaller weight
68    /// means larger error. The error has two components: the variance
69    /// of point and error in estimation of m_.
70    ///
71    void predict(const double x, double& y, double& y_err,
72                 const double w=1) ;
73
74    ///
75    /// @return prediction value and parameters
76    ///
77    std::ostream& print(std::ostream&) const;
78
79    ///
80    /// @return header for print()
81    ///
82    std::ostream& print_header(std::ostream&) const;
83
84
85  private:
86    double s2_; // noise level - the typical variance for a point with
87                // weight w is s2/w
88    double m_;
89    double m_err_; // error of estimation of mean m_
90
91  };
92
93}} // of namespace statistics and namespace theplu
94
95#endif
96
Note: See TracBrowser for help on using the repository browser.