2002 Volume 51 Issue 9 Pages 337-344
Function fitting to experimental data is one of the important procedures to derive necessary information from the data. The least-squares method has been known as a means for fitting function to normally distributed data. One of the authors (T. A.) has proposed the method for fitting to Poisson-distributed data from radioactivity measurements. We have attempted to fit functions with linear and non-linear least-squares methods and the method by T.A. and estimated the errors of the functions.
We have also obtained confidences for the intervals of the errors with computer simulations and got good results.