First, I would like to thank you for your excellent Stata programs. These have extremely helped me conduct empirical asset pricing analyses.
I was wondering if I could ask you a question about asreg.
When I run a simple regression of, for instance market excess return on stock excess return, I use the rmse option to display the root mean squared error:
asreg excret mktrf, rmse
Alternatively, I could use:
gen rmse =.
reg excret mktrf
predict res, residuals
sum res, detail
replace rmse=r(sd)
However, the outcomes for the root mean squared error differ slightly and I have no clue why.
Do you happen to have any idea what the reason behind this is?
Thank you very much in advance.
Best regards,
Phil
