Dear eva and Raj,
thanx for the comments, i have implemented your technique and it works
just fine. when i save the bootstrap samples i see how many of the samples
are greater than a specific mean using the count if command
on to another point, i am trying to do the same but with an AR(1) process,
therefore i was thinking about this command:
bs "reg var2 lag1" "e(rss)", reps(1000) saving (k:\bssample.dta)
since i want to bootstrap the residuals to generate new AR (1) processes
that could fit to the rest of the returns.
the AR(1) model is as follows:
rt= a+b r(t-1) +e
rt is returns on day t......
the question i have, is there a way after bootstraping the residuals to
fit the new models to the returns i have already to bootstrap the mean of
these returns after fitting them into the model ?? I know it is a little
bit of a complicated question, i do not know if i am asking it correctly,
but that is the way i understadn it.
Regards,
Mo
--
Mahmoud Abd El Aal
MSc. F&I
UOB
*
* 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/
