[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
re: Re: st: RE: Stata 9 announcement
I am very happy to learn that stata 9 will have linear mixed models. I
look forward to seeing the command structure for the mixed model routine,
and to seeing whether data will have to be re-structured dramatically,
minimally, or not at all.
I am also hoping that David is wrong about stata leaving it to gllamm to
work for non-linear mixed models, although this is clearly the short-term
situation. Gllamm is a godsend, but it is also very difficult to be sure
one is actually estimating what one wants to estimate--I've noticed
several queries on statalist that have that flavor. This may be because
gllamm is not integrated into stata and thus may not follow stata
conventions. I am not sure. Maybe gllamm does follow the conventions
closely, but that it is so flexible that it is difficult to document.
Whatever the reason, it seems very slow and very opaque.
At any rate, even if gllamm runs faster, nothing will substitute for
having a module for non-linear mixed models that is actually written and
supported by stata. It is my sincere hope that in a few releases (10?
11?) such will be the case. Multi-level modelling of discrete outcomes is
pretty mainstream now, for good or ill. It is my hope it will become
mainstreamed into stata 10 or stata 11 as well.
But, I am still glad to see that linear mixed models will be estimable in
stata 9. That's great news!
On Tue, 5 Apr 2005, David Airey wrote:
> > I am curious what the implications are for -gllamm-. Will it be faster
> > now? Will Stata now have built-in alternatives to much of its
> > features? I've avoided -gllamm- partly because I heard that, while it
> > was a great program, it could be unbearably slow. Speeding up -gllamm-
> > seemed to be a frequent request on this list; does V. 9 address this
> > at all?
> I'm just speculating and ignorant, but has "Mata" been put in place in
> part to deal with this? Maybe gllamm has to be re-written in Mata and
> plug-ins to achieve faster timing and maybe this is being left up to
> its authors. Stata seems to want to provide the tools to solutions in
> addition to many solutions. From the online pages, seems speed and Mata
> are in the same sentence a few times and that xtmixed is written with
> Mata. I guess there were 3 solutions to speed up gllamm, (1) code it
> into Stata (outside ado's) (2) make C plug-ins possible (3) think hard
> about the .ado architecture for programs like gllamm.
> * 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/
* For searches and help try: