You are right.
Once I open stata and generate random numbers, the seed keeps on changing
and it is completely randomized.
But since the program is long, I am running it without opening stata. I am
running it in the batch mode giving the command
stata -b do market.do
This program opens stata each time and hence gets the same seed and the same
set of random numbers each time.
Thanks,
Subh
----- Original Message -----
From: "Nick Cox" <[email protected]>
To: <[email protected]>
Sent: Thursday, May 20, 2004 10:24 AM
Subject: st: RE: Random number seed
> Your premise is false. Just because
> the seed remains the same does not
> imply identical samples (of a given size).
> Conversely, explicitly setting the seed to the
> same number as before the last
> call does ensure an identical sample.
>
> However, you can indeed change the seed easily.
>
> . search seed
>
> points to ways of doing this.
>
> More generally, there is a great
> deal of help on-line, which will
> save you time.
>
> Nick
> [email protected]
>
> > -----Original Message-----
> > From: [email protected]
> > [mailto:[email protected]]On Behalf Of Subhankar
> > Nayak
> > Sent: 20 May 2004 15:16
> > To: [email protected]
> > Subject: st: Random number seed
> >
> >
> > Can the stata random number seed be itself randomized?
> >
> > I am running a program called "market.do" in the batch mode without
> > invoking/opening stata. This program carries out an analysis
> > based on a set
> > of 1,000 uniform random numbers it generates.
> >
> > However, since the random seed is always the same everytime I run the
> > program in batch mode, I get the same set of 1,000 random
> > numbers each time
> > I run this program in batch mode. Results are identical which
> > defeats the purpose.
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
