Re: st: gllamm and mim output for multi-level model

[Stas Kolenikov <skolenik@gmail.com>]

On Mon, Oct 18, 2010 at 8:34 AM, Christian Villenas <cvillenas@gmail.com> wrote:
> Thanks so much for your response. I guess for me the confusion still
> lies in the output. If it's a matter of learning Cholesky
> decomposition, why does the gllamm output (without mim) provide the
> variances/covariances? Are you saying that I need to learn Cholesky
> decomposition in order to somehow aggregate the imputed data to get
> the  level 1 and 2 variance/covariances?

I don't know the exact mechanics, but more likely than not is that
Sophia computes the necessary delta-method derivatives directly in
-gllamm-, rather than uses Stata standard tools like -nlcom- or
-testnl-. In order for you to understand how this stuff works, you do
need to figure out Cholesky decomposition, although in your case you'd
only need to figure out how to get the covariance matrix from its
Cholesky factorization.

-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   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/