0

Is there any package in R that allows fitting many pdfs at the same time to some sample data, plots all fits and the sample histogram, and then allows sorting the fittings by some gof criteria like Kolmogorov-Smirnov, Anderson-Darling, X2, ...? Something similar to what commercial software EasyFit does?

UPDATE

I've received valuable comments to my initial question. Specifically, the AIC stands out as a metric that allows comparing pdfs with differing number of parameters. However, AIC also has limitations. Therefore, it'd be interesting to come up with/find some sort of summary that states the pros and cons of all the g.o.f. tests for model selection. Many of these topics are common to statisticians but might not be so, and are very useful for practitioners that must perform many g.o.f. decisions on a daily basis for practical problems.

Any suggestions are welcome.

Thanks!

Oliver Amundsen
  • 1,491
  • 2
  • 21
  • 40
  • Related: http://stackoverflow.com/questions/2661402/given-a-set-of-random-numbers-drawn-from-a-continuous-univariate-distribution-f/2661490 – Ben Bolker Nov 07 '14 at 20:32
  • Hi it's related but there's issues not covered in that question. Specifically, the issue of choosing a single criteria is usefully discussed but not settled as far as an R command is concerned. Thanks though! – Oliver Amundsen Nov 07 '14 at 20:51
  • I've voted to close because this is "looking for a tool". If you wanted to start to put this together for yourself I'm sure you'd get help with the details here. Also: (1) there are a relatively small number of circumstances where this is *actually* a good idea from a statistical point of view; (2) if you're fitting distributions with differing number of parameters, you'd better use some criterion like AIC that will take that into account. – Ben Bolker Nov 07 '14 at 21:00
  • Thanks for your comments. I was not familiar with using criterion for pdfs with differing number of parameters, and for that thanks!. Now two things come to mind: (1) AIC is a type of log-likelihood that penalizes no. of parameters; but insofar likelihood, it might also have its shortcomings with some distributions. Maybe such a "tool" could rank pdfs with the same number of parameters, with AIC, etc. Could you suggest a reference for such criteria? (2) I believe that this question is still useful because many people have the same errors that I had and these answers might be useful. – Oliver Amundsen Nov 10 '14 at 07:06
  • Perhaps also search on Cross Validated. – vpipkt Nov 10 '14 at 16:52

0 Answers0