A former classmate of mine from Graduate School has recently published a paper on Generalized Segmented Distributions. Utilizing this method, the distribution can approach any continuous density function of arbitrary form. The method is particularly effective in fitting data with sufficient observations that are skewed and/or multimodal. The paper can be found here:The General Segmented Distribution | Mike Vander Wielen - Academia.edu. This method solves the problem of a data set which doesn't seem to fit any distribution in the JMP menu.
I think this could and should be added to JMP. Can someone get this information into the hands of the right person at SAS?
Thank-you!
Steve
