Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Rosie Chen <jiarongchen2002@yahoo.com> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | Re: st: multinomial multilevel analysis in Stata |
Date | Mon, 17 Oct 2011 11:14:20 -0700 (PDT) |
Nick, thanks very much for helpful response. Rosie ----- Original Message ----- From: Nick Cox <n.j.cox@durham.ac.uk> To: "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu> Cc: Sent: Monday, October 17, 2011 6:06 AM Subject: RE: st: multinomial multilevel analysis in Stata This is one of the most frequent questions on this list. There are many answers, of which my favourite is to use -poisson- or -glm, link(log)- i.e. use a link function that indulges zeros in the data, not a transformation. However, if the zeros arise from some quite distinct process, you may need something more complicated. Nick n.j.cox@durham.ac.uk -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Rosie Chen Sent: 15 October 2011 15:18 To: statalist@hsphsun2.harvard.edu Subject: Re: st: multinomial multilevel analysis in Stata Thank you, Cam. I will read the articles mentioned below. What would you suggest for dealing with zero values when log transforming data? * * 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/ * * 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/