#1 #2 #3 sound highly specific to your problem; it's easy to find
fractional logit problems in which time does not even appear. But in
general yes, logit link, binomial distribution, and robust standard
errors.
For an outstandingly lucid and concise miniature review
SJ-8-2 st0147 . . . . . . . . . . . . . . Stata tip 63: Modeling proportions
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. F. Baum
Q2/08 SJ 8(2):299--303 (no commands)
tip on how to model a response variable that appears
as a proportion or fraction
http://www.stata-journal.com/sjpdf.html?articlenum=pr0032
Without re-reading it I am not sure whether he gets to the panel case,
so -xt- may come extra, so to speak.
Nick
njcoxstata@gmail.com
On 18 April 2013 07:45, samuel opoku <samuel.opoku@gmail.com> wrote:
> Dear all,
> I am new to stata and a newbie in running econometric models. I am trying to run a fractional logic model (Wooldridge & Papke, 2008, 2011) for panel data. Here is my understanding of the steps (I was hoping to get some confirmation on whether my understanding is correct please)
>
> 1. Obtain time averages for independent variables (IVs)
> 2. Include time averages as additional IVs
> 3. Include time dummies
> 4. Run a GEE model with a logic link and binomial distribution
>
> Is this correct please?
> Any help would be much appreciated.
>
> Thank you!
> Samuel
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
