For each fit, BestFit reports one or more fit statistics. These statistics measure how good the distribution fits the input data and how confident you can be that the data was produced by the distribution function. For each of these statistics, the smaller the value, the better the fit. BestFit makes use of four different fit statistics: chi-squared, Kolmogorov-Smirnov, Anderson-Darling, and Root-Mean Squared Error.
When more than one fit statistic is available, there is no hard rule to decide which test will give you the "best" result. Each test has its strengths and weaknesses. You must decide which information is most important to you when considering which test to use.
Kolmogorov-Smirinov Statistic (K-S)
Anderson-Darling Statistic (A-D)
Root-Mean Squared Error (RMSErr)
