[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

# Re: st: RE: RE: Parameter restrictions in MLE

 From Maarten buis To statalist@hsphsun2.harvard.edu Subject Re: st: RE: RE: Parameter restrictions in MLE Date Fri, 26 May 2006 23:03:30 +0100 (BST)

```Deepankar:
Your constraint is not only a linear constraint (the probabilities
should add up to one), but also a inequality constraint (all
probabilities should be larger than zero and less than one). Just
imposing the linear constraint would be equivalent to Nick's solution,
which ignores the inequality constraint.
HTH,
Maarten

--- Deepankar Basu <basu.15@osu.edu> wrote:
> Just now, I came across the "constraints" command which can be used
to
> specify linear constraints on the parameters. Since my constraints
are > linear, I can probably use this too.

> > Nick Cox
> > But one way to do it is to parameterise in terms
> > of e.g. a, b, 1 - (a + b). If you are lucky this
> > will be enough.
> >
> > You could still have problems if the algorithm
> > is happiest at e.g. a = 1.1, b = -0.9, c = 0.8.

> > > Deepankar Basu
> > > I am trying to do a maximum likelihood estimation where some of
> > > the parameters of the model are probabilities; hence they
> > > should add up to unity.

-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting adress:
Buitenveldertselaan 3 (Metropolitan), room Z214

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------

___________________________________________________________
All new Yahoo! Mail "The new Interface is stunning in its simplicity and ease of use." - PC Magazine
http://uk.docs.yahoo.com/nowyoucan.html
*
*   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/
```

 © Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index