Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | David Quinn <dxquinnx@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: "cmp" command and rho LR test statistic for SUR bivariate probit model |
Date | Thu, 3 Mar 2011 15:51:34 -0500 |
Maarten, Thanks so much for the assistance and insight on this, it has been very helpful. I have three quick follow-up questions, using the example that you applied to -cmp- in your message to me. I just want to make sure I'm understanding/interpreting the process correctly. 1.) The first question is in regards to the command: -constraint 1 [atanhrho_12]_cons = `=atanh(0)'-. When you run -cmp-, it reports the value of atanhrho_12 as -.0886333 (and this is the Fisher's z transformed correlation), and the value of rho_12 as -.088402. So, what you are doing with the -constraint- command is telling to STATA to set atanhrho_12 to a specific value when running the model? If so, does the value "0" that is in parentheses in your example represent the value of rho_12 (i.e., -.088402)? Am I understanding that correctly? 2.) The second question is in regards to the command: -est store a- and -est store b-. What exactly are these doing? I think this is where I'm getting a bit confused. 3.) To answer the third question, perhaps I first need to understand what is going on with the second question. But regardless, I'm not clear exactly what the likelihood-ratio test at the end is testing. What are you comparing with "a" and "b"? Sorry if I am a bit confused, particularly with regard to the last two questions. I think I understand everything else, though. Any insight you have would be deeply appreciated. Thanks,, --David On Mon, Feb 28, 2011 at 3:21 AM, Maarten buis <maartenbuis@yahoo.co.uk> wrote: > --- On Mon, 28/2/11, David Quinn wrote: >> Is there any way to compute the likelihood-ratio chi-square >> test statistic regarding correlation of error coefficients >> (i.e., that rho=0) when using David Roodman's "conditional >> mixed process" (aka "cmp") module to estimate a seemingly >> unrelated bivariate probit model? > > Yes, I just wrote a Stata tip for that, which will appear in > the Stata Journal 11(2). A pre-publication draft can be seen > here: <http://www.maartenbuis.nl/publications/sigma_rho.html>. > > Below is an example that applies the general idea of that tip > to -cmp- (which I got from SSC by typing in Stata > -ssc install cmp, replace- and -ssc install ghk2, replace-.) > > *------------- begin example ----------- > cmp setup > webuse laborsup, clear > replace fem_inc = fem_inc - 10 > cmp (kids = fem_inc male_educ) /// > (fem_work = male_educ), /// > ind($cmp_cont $cmp_cont) /// > quietly > est store a > > constraint 1 [atanhrho_12]_cons = `=atanh(0)' > cmp (kids = fem_inc male_educ) /// > (fem_work = male_educ), /// > ind($cmp_cont $cmp_cont) /// > quietly constraint(1) > est store b > > lrtest a b > *------------- end example ----------------- > (For more on examples I sent to the Statalist see: > http://www.maartenbuis.nl/example_faq ) > > Hope this helps, > Maarten > > -------------------------- > Maarten L. Buis > Institut fuer Soziologie > Universitaet Tuebingen > Wilhelmstrasse 36 > 72074 Tuebingen > Germany > > http://www.maartenbuis.nl > -------------------------- > > > > > * > * 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/