Look at -help simulate- or try:
capture program drop olssim
program olssim
version 10.0
drop _all
set obs 100
generate e = invnorm(uniform())*2
generate x = uniform()*10
generate y = 1 + 0.5 * x + e
regress y x
end
simulate, reps(1000) seed(123): olssim
On Tue, Apr 14, 2009 at 11:22 AM, Rudy Fichtenbaum
<[email protected]> wrote:
> I am running a monte carlo simulation using the following program:
>
> capture program drop olssim
> program olssim, rclass
> version 10.0
> drop _all
> set obs 100
> generate e = invnorm(uniform())*2
> generate x = uniform()*10
> generate y = 1 + 0.5 * x + e
> regress y x
> return scalar b0 = _coef[_cons]
> return scalar b1 = _coef[x]
> end
>
> simulate "olssim" b0 = r(b0) b1 = r(b1), reps(1000)
>
> sum b0 b1, detail
>
> The problem I am having is that each time I run the program I get a slightly
> different result. I know this is because I have not set the seed for the
> random number generator but I can't figure out how to set the seed in this
> program.
>
> Rudy
>
*
* 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/
