st: fixed effects modeling

[Andrew Reed <drew.reed1@gmail.com>]

Hi Statalisters,

I have spent quite a bit of time readying my panel data set for a fixed effects regression where I look at how credit rating agency news announcements effect the exchange rates of  different entities, 72 to be exact. 

I am wondering about the different ways to run such a regression. I've run preliminary regressions with xtreg, fe but would like to know a bit more about areg and xi:regress. I obviously want to control for any entity specific effects but I'd also like to control for time effects as well as my data set spans from 2000 to 2011. In your humble opinions, which model works best for running such a regression both in terms of estimation and in terms of generating output when I'm interested in acknowledging that entity specific effects and time effects exist but am not necessarily interested in reporting the estimates for each id value or year dummy. Xtreg seems to take care of the entity-specific effects but I'd like to hear from those that have used these differing models before.

Thanks for any help you may be able to offer as well as any perspective.

Regards,

Andrew
*
*   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/