Estimating a Fama-French Three-Factor model with Stata

[Cooper Felix]

I was reading a few papers in finance regarding the method of event study. Below is the code that I used to estimate the abnormal return of a single company, did I make any mistake?

return =stock i's return
RF =risk free return
MKT_RF, SMB, HML =variables downloaded from French's website
abnormal_return =stock i's abnormal return

gen excess = return - RF
reg excess MKT_RF SMB HML
predict phat
gen abnormal_return = return - (RF+phat)

[Cooper Felix]

any suggestions?

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output (fixed spacing fonts help), and sample data using dataex. Also, remember, we're not necessarily from your area so explain things more generally.

You need to look at the literature - there are lots of papers that say how to do this. I am not up on them, but you need to be. Without commenting on the finance issues whether this is the right way to specify a FF model, most event studies restrict the estimation sample to be some specified period before the event.