After reading through Gould et al’s excellent book on maximum likelihood
estimation in Stata, I’ve been attempting to estimate some more and less
complicated models using the d0 method. But in executing ml maximize, I
often (not always) run into the following paradox:
- The algorithm “backs up” until I get bored watching it,
suggesting some kind of problem with the likelihood function.
- Stopping the search, then using ml plot shows that it has
apparently maximized the likelihood with respect to each parameter
individually.
- The point at which it’s stuck appears to be a plausible
solution, economically.
This is driving me crazy, because I love Stata’s implementation of
maximum likelihood estimation, apart from the actual maximization bit
(attn Stata folks: an alternative ml search that uses a global
optimization method like differential evolution would be outstanding),
and in particular the fact that it apparently doesn’t want to admit it
has maximized the likelihood. I’ve tried offering initial values, and
suggesting bounds for ml search, but to no avail. Deriving gradients,
let alone Hessians, let alone coding either, is not feasible, so I’m
stuck with using d0.
*/ /*
I’m hoping that someone will have some general suggestions on how I can
diagnose the problem.*/ /*I could try loosening tolerances, but I’d
like a second opinion before I go down that road.
Thanks,
Marc Pelath
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
