st: Using stata for event studies

[lschoele@rumms.uni-mannheim.de]

Hi Statalist,

I have to do an event study. So I am thinking about using Stata. But I  
have a question about the expected returns before I start:
I set up an estimation period of 250 trading days. I use OLS to get my  
alphas and betas. How many observations of the stock price do I need  
for the regression. Would it be the best, if I have the stockprices  
for every 250 trading days? I dont really know how many observations  
would be the best for the regression.
Maybe someone can help me.

Thanks a lot
Lisa
*
*   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/