Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: ML program in Stata 12.1 for a nonlinear equation of the form Y=F(x)*G(z)


From   "JVerkuilen (Gmail)" <jvverkuilen@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: ML program in Stata 12.1 for a nonlinear equation of the form Y=F(x)*G(z)
Date   Sat, 6 Apr 2013 21:09:38 -0400

On Sat, Apr 6, 2013 at 4:35 PM, Antony Chapoto <achapoto@gmail.com> wrote:

>
> I would like your assistance to write an ML program in Stata 12.1 for
> the following nonlinear model:
>
> <snip>

This doesn't directly answer your question, but if I'm not mistaken,
this model could be estimated using -nl-, which is specialized for the
problem of nonlinear regressions with Gaussian errors, and thus has
useful post-estimation, etc. I'm a little worried that -nl- (or -ml-?)
doesn't have an option for a regularized method (such as
Levenberg-Marquardt), though, given how complicated your model is due
to the G(z) term.

That said, it looks like an unusual model with many parameters in
which homoscedastic Gaussian errors seem potentially dubious, and is
likely to have some difficulty estimating simply on numeric grounds.
I'd definitely try some pro-active simulation of likely values of the
variables.

Jay
--
JVVerkuilen, PhD
jvverkuilen@gmail.com

"It is like a finger pointing away to the moon. Do not concentrate on the
finger or you will miss all that heavenly glory." --Bruce Lee, Enter the
Dragon (1973)
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index