Statalist


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

Re: st: Variance decomposition


From   Joseph McDonnell <jockmcdock@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Variance decomposition
Date   Wed, 1 Apr 2009 12:03:26 +0930

Susan

many multilevel models are based on "location" e.g. schools within
region, students within schools, measurements within students. There
are a number of books on multilevel/random-effect models. Rabe-Hesketh
and Skrondal's Multilevel and Longitudinal Modeling Using Stata is one
that addresses such models from a Stata user's point of view and I can
personally recommend it. It's available at the StataCorp website.

Cheers

Joseph

On Wed, Apr 1, 2009 at 9:45 AM, Susan Olivia <olivia@primal.ucdavis.edu> wrote:
> Dear Stata list,
>
> I have the following model y[ch] = x’[ch]beta + u[ch]
> and u[ch] = eta[c] + e[ch],
>
> where c corresponds to location ‘c’ and h corresponds to
> household ‘h’. In other words, the error term can be
> thought of the error due to the location component and the
> household component.
>
>
> I would like to decompose the total variance due to these
> two effects. Can I get any advice how to tackle this issue?
>
> At first sight, I thought this model is similar to that of
> random effects, but we’re dealing in space rather than
> time dimension.
>
>
> Any insights on this are much appreciated. As I still have a
> lot to learn on Stata.
>
>
>
> Thanks,
>
>
>
> Susan
>
>
>
> *
> *   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/



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