Re: st: RE: comparing coefficients

["Mary E. Mackesy-Amiti" <mmamiti@uic.edu>]

Why are you not running one regression, including cohort (year) as a 
variable in the model?  Then you could include an interaction term for 
group x year.  This would tell you if the effect of group varies across 
years.


On 10/26/2010 2:35 PM, Wael Wael wrote:
> Dear all,
>
> Using STATA 11. Running a logit model for three cross sectional set
> (three years) to examine a disparity question between two groups
>
> xi: logit y i.group
>
> I ran this for each year and getting the coefficient for group 2
> (group 1 is the reference).
>
> Is there a way in STATA to compare the coefficient with each other?
>
> Thank you,
>
> W
>
> On Tue, Oct 26, 2010 at 7:31 PM, Nick Cox<n.j.cox@durham.ac.uk>  wrote:
> > Thanks for adding detail; so, now you can pose a specific question showing for example
> >
> > -logit- command,
> > -logit- output,
> > stating what is likely to be true about serial correlation in your data and what you are doing about it,
> > etc.
> >
> > I can tell you in advance that others on the list will be better placed than I am to go further down your chosen route.
> >
> > Nick
> > n.j.cox@durham.ac.uk
> >
> > Wael Wael
> >
> > Thanks for your response. I am using STATA 11.
> >
> > I used the logit syntax for each year with group 1 as a reference. I
> > would like to compare the difference i found in the first year with
> > the following year output.
> >
> > On Tue, Oct 26, 2010 at 7:00 PM, Nick Cox<n.j.cox@durham.ac.uk>  wrote:
> >
> > > There are two ways I can think of to answer this question, and many others roughly like it.
> > >
> > > 1. This is really a fishing expedition on your part in which you seek ideas on what to do in a project as yet undefined. That kind of question is not disallowed, and you may get some useful advice, but on the whole your chance of getting good advice would be much increased if you gave much more detail.
> > >
> > > 2. You are already using some specific commands. That kind of question is more likely to get answers, but again only with supply of Stata detail. For example, you don't even say what model or models you are using or contemplating.
> > Wael Wael
> >
> > > I am running a serial cross sectional study to examine disparities
> > > between two groups. I want to compare my model output in year 1 with
> > > the following year. (i.e. is the gap in disparities increasing or
> > > decreasing)
> > >
> > > Any ideas on how to do this in STATA would be appreciated.
> > *
> > *   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/

--
Mary Ellen Mackesy-Amiti, Ph.D.
Research Assistant Professor
Community Outreach Intervention Projects (COIP)
School of Public Health m/c 923
Division of Epidemiology and Biostatistics
University of Illinois at Chicago
ph. 312-355-4892
fax: 312-996-1450

*
*   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/