Fitting Log Pearson Type III distribution in MATLAB

I need to (1) fit Log Pearson Type III distribution to a dataset and estimate the parameters; and (2) create a probability distribution object using the parameters estimated in task 1. I know that for other distributions (e.g., GEV), 'fitdist' and 'makedist' can be used to accomplish task 1 and 2, respectively; but these cannot be applied for Log Pearson Type III. Would someone kindly help me in this regard? I will greatly appreciate any kind of advice/suggestion. Thank you.

Answer 1

[Jeff Miller]

I think you can make a Log Pearson III distribution and estimate its parameters with the routines in Cupid . The attached demo shows how.

I am not sure what you want to do with the probability distribution object after you create it; maybe the Cupid routines will do some or all of that as well. But Cupid's probability distribution objects are a little different from MATLAB's, so also maybe not.

HTH