Xref: utzoo comp.graphics:14149 sci.math:13143
Newsgroups: comp.graphics,sci.math
Path: utzoo!utgpu!news-server.csri.toronto.edu!helios.physics.utoronto.ca!alchemy.chem.utoronto.ca!mroussel
From: mroussel@alchemy.chem.utoronto.ca (Marc Roussel)
Subject: Re: question regarding random numbers
Message-ID: <1990Oct31.035424.27431@alchemy.chem.utoronto.ca>
Organization: Department of Chemistry, University of Toronto
References: <29358@pasteur.Berkeley.EDU> <CEDMAN.90Oct29210857@lynx.ps.uci.edu> <6936@castle.ed.ac.uk>
Date: Wed, 31 Oct 90 03:54:24 GMT

In article <6936@castle.ed.ac.uk> aipdc@castle.ed.ac.uk (Paul D. Crowley) writes
>cedman@lynx.ps.uci.edu (Carl Edman) writes:
>>There is one general method of getting random variables given an arbitrary
>>distribution. It is known under the handy name of Metropolis-Rosenbluth-
>>-Rosenbluth-Teller-Teller-algorithm.
>
>Is that for real?  Excuse me while I crosspost to rec.humor... anyone
>know how it got such a wonderful name?

     At least in the Chemistry community, the technique is usually simply
known as Metropolis sampling.  I have heard it called rejection sampling
in other places.  The idiosyncratic MRRTT name which Carl gave it comes
from the list of authors on the paper in which the method was first
described.  (I can post the reference if anyone is interested.)
Apparently (so the story goes), Metropolis, the Rosenbluths and the
Tellers developed the method on a paper napkin at a dinner party.

				Marc R. Roussel
                                mroussel@alchemy.chem.utoronto.ca