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

From |
Partha Deb <partha.deb@hunter.cuny.edu> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: question with MLE and complex model |

Date |
Mon, 04 Aug 2008 11:25:39 -0400 |

.

Stas is correct that the problem does seem odd. But without specifics, I'm having a hard time with diagnoses.

-iter()- and -from()- both work with -fmm- . The mixing probabilities are, in fact, specified as multinomial logits.

clear all

use http://urban.hunter.cuny.edu/~deb/Stata/datasets/mepssmall.dta

gen mc=(instype>1)

fmm docvis mc female minority age nchroniccond, mi(negbin2) co(2)

fmm docvis mc female minority age nchroniccond, mi(negbin2) co(2) ///

pr(female) iter(0)

does what Stas is suggesting.

Partha

Stas Kolenikov wrote:

On Thu, Jul 31, 2008 at 2:22 PM, Jonathan Hanson <johanson@maxwell.syr.edu> wrote:What happens when you estimate 2 and 3 component mixtures without covariates in the probabilities?I can get results with no covariates in the probabilities when starting with a 2 component mixture and then moving to 3 component mixture. Usually, as soon as I add one covariate in the probabilities, Stata cannot find initial values, whether there are 2 or three components.That's odd -- if you specify the value of zero for your covariate, then this is precisely the model with fixed probabilities. If the latter converged, then at least for one set of values the likelihood is computable! And most likely in the vicinity of that model the likelihood should be well defined. Of course Stata might be making big jumps with the -ml search- procedure getting out of the domains of reasonable estimates for which the likelihood is computable. It depends on the implementation details though, and I am drawing a blank about how the mixture probabilities were coded -- a multinomial logit, I would think (Partha?) A crude way of looking into what -from- is expecting is to supply a vector of zeroes or ones as the -from- option, and also specify -iter(0)- option (I hope Partha's code supports transfer of maximization options; this is good -ml- programming style!). Then the estimation will conclude without starting, and you can look into the format of the parameter vector using mat li e(b)

-- Partha Deb Department of Economics Hunter College ph: (212) 772-5435 fax: (212) 772-5398 http://urban.hunter.cuny.edu/~deb/ Emancipate yourselves from mental slavery None but ourselves can free our minds. - Bob Marley * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**References**:**Re: st: question with MLE and complex model***From:*"Stas Kolenikov" <skolenik@gmail.com>

- Prev by Date:
**st: RE: RE: list command** - Next by Date:
**RE: st: RE: Count no of unique occurrences on a prior-12-month basis** - Previous by thread:
**Re: st: question with MLE and complex model** - Next by thread:
**Re: st: Heteroscedastic Tobit** - Index(es):

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