"Chia C Chong" <[EMAIL PROTECTED]> wrote in message
ah3mj9$lpj$[EMAIL PROTECTED]">news:ah3mj9$lpj$[EMAIL PROTECTED]...
> Hi!
> I read some stochastic books on B-D process. Most of them stated that
> "birth" means increase a population (for e.g.) by "1" and "death" means
> decrease a population by "1". My question is, if for a particular time
> interval, I have more than 1 death & 1 birth, does this implies that I
can't
> apply the simple B-D process for my case? If yes, what are the more
> appropriate stochastic process that I need to look into?
>
> Thanks so much/
>
>
> Regards,
> CCC
>
>
>
Do a search on compound Poisson processes. Interarrival times are negatively
exponentially distributed, but the number of arrivals can be a scalar or
even a vector. Similar, but more complicated processes can be used to model
deaths. Do a search on batch service queues.

--
Good luck,

Jerry Harder
remove spamnein from address to reply


.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to