In article <[EMAIL PROTECTED]>,
Rich Ulrich <[EMAIL PROTECTED]> wrote:
>On 23 Nov 1999 09:16:42 -0800, [EMAIL PROTECTED] (Bob Hayden)
>wrote:
>> Here is a query from someone I met in a hallway a couple weeks ago.
>> I'm not sure I even understand the question. If anyone out there
>> recognizes anything familiar in this scenario, you might respond to
>> JMC -- not to me or the list.
> - well, here is a short comment, anyway... also e-mailed, since Forte
>Agent can do that quite conveniently.
>> ----- Forwarded message from Jennifer Mary Collins -----
>> The problem is that the data I use, which is hurricane numbers, deals
>> with small numbers. Hence they are poisson distributed, when one
>> considers the other criteria to be poisson.
> You might as well say, Hurricanes disrupt little fishes and therefore
>the distribution is poisson (== "fishes" in French). -- Nonsequitur.
>Having small numbers has little or nothing to do with being Poisson.
>Having *counts* is where Poisson usually comes in, but the example
>which is given doesn't show a count, anywhere.
Even having counts does not make something Poisson, or even
approximately Poisson. If there are other things affecting
the number of hurricanes in a given year, even if the number
given these random variables is Poisson with a mean for that
year, the yearly numbers will not be Poisson. For example,
if the yearly means have a Gamma distribution, the distribution
of the numbers of hurricanes will be negative binomial.
But it is not hard to get models where hurricanes have some
lack of independence in a given year. This can produce
distributions which are not even compound Poisson. It seems
that the assumptions are going to make a big difference in
the way that you look at the data, so use your knowledge of
the physical process to make a realistic model using the
knowledge of geophysical and meteorological processes, not
inappropriate probability assumptions for convenience.
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
[EMAIL PROTECTED] Phone: (765)494-6054 FAX: (765)494-0558
