[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Any randomly generated numbers according to *non-uniform* probabili
From: |
donalson |
Subject: |
Re: Any randomly generated numbers according to *non-uniform* probability distribution? |
Date: |
Fri, 16 Apr 1999 04:34:01 -0700 |
Check out the book "Simulation Modeling and Analysis" by Law and Kelton
(McGraw/Hill). It has a lot of random distrubutions defined as well as a good
section on how to code them. Sorry it took a while for this reply but the book
had hidden itself when I wasn't looking and I just found it again.
Cheers,
Doug Donalson
"William S. Shu" wrote:
> I would like to attribute the probablity of events occuring based on known
> (or user-defined probability distributions other than "uniformly random".
> E.g., An agent has different probabilities of being in one of a number of
> distinct positions, and has to visit it based on this probability. The way
> I currently implement it is associate each position with an interval of the
> real line (in the range [0,1]), the length of which corresponds to its
> probablility of being visited. I then randomly generate numbers in [0,1].
> If the number falls in the line segment corresponing to a position, I move
> to that position.
>
> The trouble is that I have a number of such scenarios, for different
> situations with different probabilities. I cannot easily build them,
> especially for more complicated cases.
>
> Do packages exist to do this, or similar?
>
> In practical terms, could someone tell me if there are any random number
> generators whose output over an interval [a,b] is generated with a
> probability distributed according to common (known) or user-defined
> probability distributions. (Common distributions would include, say,
> normal, binomial, distributions). As a (crude) illustration, suppose such a
> generator is producing numbers over the interval [-1.0,1.0] according to a
> Normal Probability Distribution. Then, it will produce any number in the
> range [-1.0,0.0] with a probability of 0.5.
>
> Thanks
>
> William.
>
> ==================================
> Swarm-Support is for discussion of the technical details of the day
> to day usage of Swarm. For list administration needs (esp.
> [un]subscribing), please send a message to <address@hidden>
> with "help" in the body of the message.
--
*********************************************************************
* Doug Donalson Office: (805) 893-2962
* Ecology, Evolution, Home: (805) 961-4447
* and Marine Biology email address@hidden
* UC Santa Barbara
* Santa Barbara Ca. 93106
*********************************************************************
*
* The most exciting phrase to hear in science, the one that
* heralds new discoveries, is not "EUREKA" (I have found it) but
* "That's funny ...?"
*
* Isaac Asimov
*
*********************************************************************
==================================
Swarm-Support is for discussion of the technical details of the day
to day usage of Swarm. For list administration needs (esp.
[un]subscribing), please send a message to <address@hidden>
with "help" in the body of the message.