Tiago V. Pereira <[email protected]>, in response to my previous
response to his question on resetting the random number seed, writes,
> I thought it would be appropriate to add "a new pseudo-seed" inside
> the loop because I need to do a loop, say, with 1000 replications,
> in which each replication has then it own 1000 replications (e.g.
> a loop inside a loop). [...] That´s why my message is "how to get
> random seeds more efficiently?". Any other suggestions?
In my previously repsonse I addressed exactly that issue. Stripped
of explanation, the operational advice was (1) set the seed once outside
of all loops, (2) do *NOT* reset the seed inside the inner loop, and
(3) if you wish to be able to reproduce individual inner results
in isolation, include -display "`c(seed)'"- at the top of the inner loop.
-- Bill
[email protected]
*
* 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/
