Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: is gllamm appropriate? is it necessary?


From   Michael Norman Mitchell <Michael.Norman.Mitchell@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: is gllamm appropriate? is it necessary?
Date   Sat, 13 Mar 2010 12:48:34 -0800

Dear Jessi

You have articulated your problem very well, but I think that we would need to know just a little bit more to give you a good answer.

  1. In a couple of sentences, what is your research question.
2. What role does "county" play? (Is it just a control variable? Is it a clustering variable? Is it a potential variable you want to interact with the other variables?) 3. How do you want to treat the "cause of death" variable. Do you want to model all 9 levels simultaneously? If so, each "cause" would be modeled against a "reference cause". Or do you want to model each cause of death against all others (e.g. cause 1 vs. all others. cause 2 vs. all others, etc.)

I think that, based on this information, people might have thoughts on the statistical model that would answer your question, which then would indicate what commands would be appropriate for answering your research question(s).

Michael N. Mitchell
See the Stata tidbit of the week at...
http://www.MichaelNormanMitchell.com

On 2010-03-13 11.23 AM, Jessica Bishop-Royse wrote:
Hello Statalisters!  Hope you're having a wonderful weekend.

I have a small question regarding multilevel modeling.

I am working with a dataset with over 300,000 observations for two years of
data.  I am trying to model a 9 category nominal dependent variable (cause
of death), my independent variables (there are 14, which are continuous and
dichotmous) as well as 6 county level indicators.

My question this:  is this best done with gllamm?  Does anyone have any
other suggestions for a more appropriate model?  Or perhaps a suggestion as
to how to speed this up?

Any help or suggestions are much appreciated!

Thanks so much, in advance.

Jessi Bishop-Royse

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index