Re: Ask Dr. Random

[Sven N. Thommesen]

Hey Steve (whether I know more about probability than you do, or not):

I am painfully aware that the set of random distributions available in the 
Swarm random library is (ahem) somewhat skimpy. As soon as I finish a 
current little after-hours project I'm working on, I shall turn my 
attention to remedying that deficiency.

In the mean time, one place you can go to find distributions is the GSL 
library, which can be found at

<http://sourceware.cygnus.com/gsl/> or, as it is known these days, 
<http://sources.redhat.com/gsl/>.

Two places to go for general info on random issues:

<http://www.ncsa.uiuc.edu/Apps/CMP/PRNG/www-rng.html>
<http://random.mat.sbg.ac.at>

Hope that was at least a little helpful,
Sven

At 01:38 PM 11/15/00, you wrote:
> Hey Sven (or anybody else that knows more about probability than I do):
>
> I have a problem that I think would best be modeled using a Poisson
> distribution, which is not in the Random library.
>
> The mean number of times per day a single fish is caught by fisherpeople
> is 0.5 (for example; there are a LOT of people fishing for hatchery
> trout at this site). I need to randomly draw the number of times each
> individual fish was caught in a day, which will be an integer of zero or
> greater.
>
> I believe this problem is best represented with the Poisson
> distribution, which models the number of times an event occurs within a
> specified time period, given the mean occurrence rate. (The exponential
> distribution would tell me the probability of a fish being caught at
> least once during a day, which is not what I want.)
>
> This seems like a fairly common kind of problem. Has somebody
> implemented it already?
>
> Steve
> --
> address@hidden
> Lang, Railsback & Assoc.
> 250 California Ave., Arcata CA 95521
> 707-822-0453; Fax 822-1868
>
>
>                   ==================================
>    Swarm-Modelling is for discussion of Simulation and Modelling techniques
>    esp. using Swarm.  For list administration needs (esp. [un]subscribing),
>    please send a message to <address@hidden> with "help" in the
>    body of the message.
>                   ==================================



                 ==================================
  Swarm-Modelling is for discussion of Simulation and Modelling techniques
  esp. using Swarm.  For list administration needs (esp. [un]subscribing),
  please send a message to <address@hidden> with "help" in the
  body of the message.
                 ==================================