I'm not Andy. And I disagree -- it might provide an equally sensible
objective function. log-normal and poisson can be similar for large
means, which can be faked via scaling, depending on the data and the
problem.
Of course, it all depends, as usual.
On 9/3/06, Spencer Graves <spencer.graves@pdf.com> wrote:
Hi, Andy:
Correct. However, it should at least get her a more sensible
objective function.
Spencer Graves
A.J. Rossini wrote:
> But that wouldn't necessarily do nonlinear regression.
>
> On 9/3/06, Spencer Graves <spencer.graves@pdf.com> wrote:
>> Have you considered 'glm(... family=poisson)'?
>>
>> Hope this helps.
>> Spencer Graves
>>
>> Paternostro.Amy@epamail.epa.gov wrote:
>> > I am currently running a script in S-PLUS 2000 on a Windows platform
>> > using the canned S-PLUS nonlinear regression function nls. I was
>> > wondering if anyone on the list has had any experience creating their
>> > own nonlinear regression function using something other than least
>> > squares? I am particularly interested in utilizing a Poisson family in
>> > my calculations.
>> >
>> > Thanks,
>> > Amy Paternostro
>> >
>> > --------------------------------------------------------------------
>> > This message was distributed by s-news@lists.biostat.wustl.edu. To
>> > unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
>> > the BODY of the message: unsubscribe s-news
>> >
>> --------------------------------------------------------------------
>> This message was distributed by s-news@lists.biostat.wustl.edu. To
>> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
>> the BODY of the message: unsubscribe s-news
>>
>
>
--
best,
-tony
blindglobe@gmail.com
Muttenz, Switzerland.
"Commit early,commit often, and commit in a repository from which we can easily
roll-back your mistakes" (AJR, 4Jan05).
|