|
|
|
|
Subject: RE: cormas random generated around a mean/SD?
From: Christophe LE PAGE (christophe.le_page@cirad.fr)
Date: Tue Feb 16 2010 - 08:05:44 CET
Hi Deb,
Don’t you have any other information about the distribution function ?
Is it a log-normal distribution? If yes, you may consider this:
The random variable $$\left( {\prod {_{i = 1}^n {{X_i } \mathord{\left/
{\vphantom {{X_i } {X_{i + n} }}} \right. \kern-\nulldelimiterspace} {X_{i +
n} }}} } \right)^{{1 \mathord{\left/ {\vphantom {1 {\sqrt {2n} }}} \right.
\kern-\nulldelimiterspace} {\sqrt {2n} }}}$$
can be used to generate standard log-normal variables Lambda(0, 1), where
the Xi are independent uniform variables on [0, 1].
With Cormas you can easily get uniform variables on [0, 1] by using “Cormas
random”
But maybe someone already implemented a method to generate log-normal
distributions?
HTH,
clp
De : owner-cormas@cirad.cirad.fr [mailto:owner-cormas@cirad.cirad.fr] De la
part de Deb Cleland
Envoyé : mardi 16 février 2010 04:58
À : Cormas
Objet : cormas random generated around a mean/SD?
Hi all,
Another potentially silly question from an amateur.
Can I generate a random number around a mean and standard deviation? Eg if I
have a distribution which ranges from 10 to 2000, with an average of 86, is
it possible to generate a random number between those amounts that will
result in a similar distribution?
Thanks very much again in advance for your help.
Deb
--
Deborah Cleland
PhD Student
Fenner School of Environment and Society
College of Medicine, Biology & Environment
Room 2.05, Building 48a
Australian National University
Canberra, A.C.T. 0200
p: +61 2 6125 8150
m: (fil) +63 9 155 074 050
(aus) +61 408 283 852
e: deborah.cleland@anu.edu.au
<<<- Is it worth a tree to print me?
"Every morning I awake torn between a desire to save the world and an
inclination to savor it. This makes it hard to plan the day." E.B. White
