Re: st: run cross-sectional regression for a subset of panel data

[Richard Goldstein <richgold@ix.netcom.com>]

have you looked at the -statsby- command?

Rich

On 11/6/13, 4:35 PM, Huanhuan Shi wrote:
> Dear Statalist,
> 
> I've been tried to solve this problem with a few lines of command but
> failed. Could you please give me some suggestions?
> The problem is: I have panel data with firms and years and I have a
> variable called sales. I want to create a new variable called coeff.
> The way to get coeff is: For firm i at time t, run a regression using
> firm i's past 5 years sales as dependent variable and use 5, 4, 3, 2,
> 1 as independent variable. For this cross sectional regression with
> only 5 datapoints, I want to collect the coefficent of sales as the
> new variable coeff. By repeating this procedure for each firm i at
> time t, coeff is created.
> 
> Thank you so much for your help!!!!!
> 
> Best,
> 
> Lucy
*
*   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/