Re: st: Re: psedual R2 of mim:gllamm fitted models

[jl591164@albany.edu]

Thanks to Stas and Lachenbruch.

Follwing the advice, does this mean that I may need to compare the
sensitivity and specificity of the two models? How i may caculate or
obtain the sensitivity and speficicity after fitting the two gllamm
models?

gllapred p, p yields the probabilities of the response variable at each of
the five study waves. Should i use the command of gallapred, p, or the
gllapred, mu?

What is a good validation sample? I would think that one of the multiple
imputed dataset would be a cadidate. An alternative would be the original
dataset before multiple imputation. However, due to large missing data
across the response and the covariate varibles, i may not get a sample of
a reasonble sample size, with complete data on all study variables.

Thanks in advance for any advice.

Jun Qing


 > On Fri, May 7, 2010 at 11:52 AM, Lachenbruch, Peter
> <Peter.Lachenbruch@oregonstate.edu> wrote:
>> In addition, the interpretation as 'explained variance' is
>> inappropriate.  The name pseudo-R2 leads the unwary to this
>> interpretation and we need to be cautious.  Of course, this list doesn't
>> need this warning.
>
> Obviously, I do :))
>
> --
> 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/
>

*
*   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/