Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | "JVerkuilen (Gmail)" <jvverkuilen@gmail.com> |
| To | statalist@hsphsun2.harvard.edu |
| Subject | Re: st: Using natural logs on RHS of maximum likelihood models |
| Date | Wed, 27 Mar 2013 00:41:53 -0400 |
On Tue, Mar 26, 2013 at 10:18 PM, David Hoaglin <dchoaglin@gmail.com> wrote: someone else reading this list may be able > to recommend a book that discusses this and other steps in the > model-building process and shows how they work on actual sets of data. > A. C. Atkinson. 1986. Plots, Transformations and Regression: An Introduction to Graphical Methods of Diagnostic Regression Analysis. Oxford. Has a chapter on GLMs. (Not coincidentally the book cites work by David Hoaglin.) Anyway, the idea behind log-transforming an RHS variable would be to reduce high leverage points induced by a long right tailed distribution such as income, wage, or something like that. It poses no in principle problems but of course a log-transformed variable is not the same as the oroginal variable. * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/