st: Including fixed effects in a panel regression

[Krzysztof Gajewski <kgnbp1@gmail.com>]

Dear Statalist Users,

I want to estimate a model to capture the impact of some variables on
bank's credit supply using Stata 11. My data-set has a following panel
structure:
date, bank, firm, credit, set of explanatory variables (specific to the bank).

In the sample, I have 24 quarters, around 150 different banks and over
50 thousand different firms. I use "xtset bank date" to set the panel
and I want to run the following panel regression for bank i and firm
j*:

(dcredit)ij= Bj+B1(explanatory_variables)i+eij,
where (dcredit)ij is change of credit granted to firm j by bank i,
(explanatory_variables)i include set of bank's i characteristics, and
Bj are firm's j fixed effetcs.

My question is: how to include firm's fixed effect in a regression
where my panel variable is bank (and I have over 50 thousand different
firms)?

* Similar model is used in Khwaja and Mian "Tracing the Impact of Bank
Liquidity Shocks: Evidence from an Emerging Market", American Economic
Review 2008, 98:4, 1413–1442,
http://www.aeaweb.org/articles.php?doi=10.1257/aer.98.4.1413

Thanks for your consideration

Best regards,

Christopher

*
*   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/