Free cookie consent management tool by TermsFeed Policy Generator

source: branches/StatisticalTesting/HeuristicLab.Analysis.Statistics/3.3/LinearLeastSquaresFitting.cs @ 9706

Last change on this file since 9706 was 9706, checked in by ascheibe, 11 years ago

#2031

  • added exponential fitting
  • added logarithmic fitting
  • refactored fitting code
  • updated license headers
File size: 3.1 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2013 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
4 *
5 * This file is part of HeuristicLab.
6 *
7 * HeuristicLab is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU General Public License as published by
9 * the Free Software Foundation, either version 3 of the License, or
10 * (at your option) any later version.
11 *
12 * HeuristicLab is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
15 * GNU General Public License for more details.
16 *
17 * You should have received a copy of the GNU General Public License
18 * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
19 */
20#endregion
21
22using System;
23using System.Linq;
24
25namespace HeuristicLab.Analysis.Statistics {
26  public class LinearLeastSquaresFitting : IFitting {
27    public void Calculate(double[] dataPoints, out double p0, out double p1) {
28      var stdX = Enumerable.Range(0, dataPoints.Count()).Select(x => (double)x).ToArray();
29      Calculate(dataPoints, stdX, out p0, out p1);
30    }
31
32    public void Calculate(double[] y, double[] x, out double p0, out double p1) {
33      if (y.Count() != x.Count()) {
34        throw new ArgumentException("The lenght of x and y needs do be equal. ");
35      }
36
37      double sxy = 0.0;
38      double sxx = 0.0;
39      int n = y.Count();
40      double sy = y.Sum();
41      double sx = ((n - 1) * n) / 2;
42      double avgy = sy / n;
43      double avgx = sx / n;
44
45      for (int i = 0; i < n; i++) {
46        sxy += x[i] * y[i];
47        sxx += x[i] * x[i];
48      }
49
50      p0 = (sxy - (n * avgx * avgy)) / (sxx - (n * avgx * avgx));
51      p1 = avgy - p0 * avgx;
52    }
53
54    //TODO: adapt or remove
55    public double CalculateError(double[] dataPoints, double a1, double a0) {
56      double r = 0.0;
57      double avgy = dataPoints.Average();
58      double sstot = 0.0;
59      double sserr = 0.0;
60
61      for (int i = 0; i < dataPoints.Count(); i++) {
62        double y = a1 * i + a0;
63        sstot += Math.Pow(dataPoints[i] - avgy, 2);
64        sserr += Math.Pow(dataPoints[i] - y, 2);
65      }
66
67      r = 1.0 - (sserr / sstot);
68      return r;
69    }
70
71    public DataRow CalculateFittedLine(double[] y, double[] x, string rowName) {
72      double k, d;
73      Calculate(y, x, out k, out d);
74
75      DataRow newRow = new DataRow(rowName);
76      for (int i = 0; i < x.Count(); i++) {
77        newRow.Values.Add(k * x[i] + d);
78      }
79      return newRow;
80    }
81
82    public DataRow CalculateFittedLine(double[] dataPoints, string rowName) {
83      DataRow newRow = new DataRow(rowName);
84      double c0, c1;
85      Calculate(dataPoints, out c0, out c1);
86      var stdX = Enumerable.Range(0, dataPoints.Count()).Select(x => (double)x).ToArray();
87
88      for (int i = 0; i < stdX.Count(); i++) {
89        newRow.Values.Add(c0 * stdX[i] + c1);
90      }
91
92      return newRow;
93    }
94
95    public override string ToString() {
96      return "Linear Fitting";
97    }
98  }
99}
Note: See TracBrowser for help on using the repository browser.