Fit/Auto::Fit

The Auto::Fit command can be accessed in the Fit menu and by the Auto::Fit button on the Speed Bar.

The Auto::Fit function will automatically choose appropriate continuous or discrete distributions to fit to the input data, calculate Maximum Likelihood Estimates for those distributions, test the results for Goodness of Fit, and display the distributions in order of their relative rank.   The relative rank is determined by an empirical method which uses effective goodness of fit calculations.   While a good rank usually indicates that the fitted distribution is a good representation of the input data, an absolute indication of the goodness of fit is also given.

This acceptance of fit usually reflects the results of the goodness of fit tests at the level of significance chosen by the user.   However the acceptance may be modified if the fitted distribution would generate significantly more data points in the tails of the distribution than are indicated by the input data.

For continuous distributions, the Auto::Fit dialog can limit the number of distributions by choosing only those distributions with a lower bound or by forcing a lower bound to a specific value as in Fit Setup.   Also, the number of distributions will be limited if the skewness of the input data is negative; many continuous distributions with lower bounds do not have good parameter estimates in this situation.

For discrete distributions, the Auto::Fit dialog limits the distributions by choosing only those distributions that can be fit to the data.   The discrete distributions must have a lower bound.

The Auto::Fit function forces the setup of the document so that only continuous or only discrete distributions will be used, both the Kolmogorov Smirnov test and the Anderson Darling test (Chi-Squared test for discrete distribution) will be calculated, the Maximum Likelihood Estimates will be used, and the accuracy will be at least 0.0003.   Because the Auto::Fit function requires a specific setup, the Auto::Fit view will be closed if the step is changed.

The manual method of fitting distributions can be used instead of Auto::Fit.