Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Adriaan Hoogendoorn <aw.hoogendoorn@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: How to model a positive continuous dependent variable with many zeros? |
Date | Thu, 9 Jun 2011 15:24:52 +0200 |
Dear Steve, Hitesh and Maarten, Steves idea of modelling proportions put us on the right track, because we figured out that these proportions are the important outcome variable here. The reference to Kit Baum's Stata Tip was very useful, since it explains how proportions with zeroes (and ones) can be modelled by a GLM with a logit link function and a binomial distribution as family choice. It turned out to be possible to extend the GLM by use of GLLAMM to incorporate the multilevel structure. Thank you for suggesting the use of two-part models for dealing with the many zeroes, which may become usefull in future problems. Kind regards, Adriaan Hoogendoorn GGZ inGeest, Amsterdam * * 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/