| Re: numerical recipes (was xerr) [message #64354] |
Thu, 18 December 2008 10:43 |
![Paul Van Delst[1] is currently offline](theme/default/images/xoffline.png.pagespeed.ic.XRkd1fkXye.png "Paul Van Delst[1] is currently offline")
Paul Van Delst[1]
Messages: 1157
Registered: April 2002
|
Senior Member |
|
|
Paolo wrote:
>
> Here goes another tangent remarks:
>
> I have always been a bit worried about the lack of
> real documentation or source code for the IDL random
> functions...
>
> Since that's so important for the montecarlo stuff
> etc., I think we have a right to know what goes
> on under the hood! I mean, if a referee complains
> about your random number generator, how can
> you answer? You'll answer, I am using "something
> similar to ran1()" (to quote the docs)?
A similar question comes up in the Fortran newsgroups every now and again. The general
response is usually "if you depends on your random number generator, then you should
implement it yourself".
> So, has anybody implemented instead one of the
> new (and better) algorithms proposed in the third
> edition of NR?
> They seem to be pretty good to me.
The Mersenne Twister algorithm comes up a lot in clf. But I think even that one has issues
(details escape me right now....)
> I think this is one very important point, but then
> maybe I am worrying too much ;-)
If you do Monte Carlo simulations you should be worried about your RNG. The fact that you
actually do worry about it indicates you're a Good Scientist (tm). :o)
cheers,
paulv
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
