Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Nick Cox" <n.j.cox@durham.ac.uk> |
To | <statalist@hsphsun2.harvard.edu> |
Subject | RE: st: How to test whether data follows Exp distribution? |
Date | Wed, 7 Jul 2010 19:44:55 +0100 |
As I understand it, -reldist- could be useful with preliminary work. But other approaches suggested to date are simpler and more direct for the question asked. While I am it, Tony's suggestion to me makes most sense if subsequently you are going to work with a log transform of the variable. Both procedures he mentioned are very indulgent of irregularities in the far tail. That may be exactly appropriate, or the opposite, depending on what follows. Nick n.j.cox@durham.ac.uk Abhimanyu Arora Just out of curiosity, can Ben Jann's -reldist- command be useful for the purpose? On Wed, Jul 7, 2010 at 8:01 PM, Lachenbruch, Peter > If I recall my basic math stat, log(Y) should be uniform if Y is exponential. So you could check it that way. A very simple idea would be to bin the logs and do a chi-squared gof test. Maarten buis > --- On Wed, 7/7/10, Jabr, Wael M wrote: >> I am trying to find if the variable I >> have follows an exponential distribution. > > You can use -hangroot- to check an empirical > distribution against, among others, an > exponential distribution. To install it type > in Stata -ssc install hangroot-. > * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/