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: Adjustment to likehood value due to dependence of data observations


From   Nick Cox <njcoxstata@gmail.com>
To   Abdul Q Memon <a.memon@ucl.ac.uk>
Subject   Re: st: Adjustment to likehood value due to dependence of data observations
Date   Thu, 22 Sep 2011 16:25:39 +0100

Answers below. Please note the advice on personal answers to Statalist
posters given in the FAQ>

On Thu, Sep 22, 2011 at 4:08 PM, Abdul Q Memon <a.memon@ucl.ac.uk> wrote:
> Dear Nick
>
> Thanks for a quick reply. Just to clearify few points.
>
> 1. As GEE takes a long time to run (quarter of million observations) so I
> decided to run GLM which is quite fast and compared the likelihood values
> for model selection. Based on those likelihood values I selected the model
> and used the GEE for that. Now the question has arised that since it is
> panel data, so all my likelihood values (obtained by using GLM) are wrong
> due to dependent obervations. I am looking for a way to correct
> (correction factor) for the likelihood to adjust the likelihood for glm
> which is used for model selection.

I see what you are doing, but my advice is essentially the same. The
likelihood can't just be adjusted unless you know the right answer any
way. Whether you can devise some approximation for cases where you
know both results I don't know.

> 2. After running the GEE which is suitable for model (Panel data) it seems
> there is still some trend in residuals (Plot of residuals and fitted
> values possibly becuase of homoscadicity). Am i right in using robust
> after gee command?? I have looked at adding some more variables but afraid
> that have no new ones.
>
> Some more input from you might clear up what I am missing.

This is the same question pretty much, and my answer is the same, more
or less. You have structure related to time. Having no more variables
is not a problem here. You need a better model for how your process
works in time. I have no idea what that model could be.

However, you do seem a bit confused. Trend and homoscedasticity [NB
spelling for your dissertation/paper/report] are different things.
Trend is often interesting but problematic, but it all depends on what
the model is. Homoscedasticity is usually good news, but not always
(e.g. if you think the process is Poisson). It is possible for trend
and heteroscedasticity to be tangled up together, but that is a
different story.
*
*   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