This algorithm for the least squares fit uses a logarithmic abscissa, and tries to find an equation matching the data set with the form y = M * log(X)/log(10) + B, using the least squares method. mybloggingplanet.com
The function below finds M and B, and if no solution exists, it returns 0 for both these values.
public static void LeastSquaresFitLogAbscissa(Pnt[] points, int numPoints, ref double M, ref double B) { //Gives best fit of data to curve Y = M*log(X)/log(10) + B double x1, y1, xy, x2, J, LX; int i; x1 = 0.0; y1 = 0.0; xy = 0.0; x2 = 0.0; for (i = 0; i < numPoints; i++) { LX = Math.Log10(points[i].X); x1 = x1 + LX; y1 = y1 + points[i].Y; xy = xy + points[i].Y * LX; x2 = x2 + LX * LX; } J = ((double)numPoints * x2) - (x1 * x1); if (J != 0.0) { M = (((double)numPoints * xy) - (x1 * y1)) / J; M = Math.Floor(1.0E3 * M + 0.5) / 1.0E3; B = ((y1 * x2) - (x1 * xy)) / J; B = Math.Floor(1.0E3 * B + 0.5) / 1.0E3; } else { M = 0; B = 0; } }
Comments