Re: st: RE: Membership (component) Determinant using FMM Command

[Partha Deb <partha.deb@hunter.cuny.edu>]

Allen,

After you have parameter estimates from -fmm-, you can get predictions 
of various types using the following syntax:

predict predvar if e(sample), equation(component#)

where predvar is the predicted variable name and component# is 
component1, component2, ...

or predictions of posterior probabilities using

predict predvar if e(sample), post equation(component#)

Once you have the posterior probabilities, you can use the ranking of 
the posterior probabilities to assign an observation into a class.  In 
the 2 component case, this is equivalent to checking whether the 
posterior probability of being in component1 is > (<) 0.5.

Hope this helps.

Partha

On 2/16/2011 11:46 AM, Nick Cox wrote:
> This is the third time of asking and goes beyond Statalist etiquette on this point, which is clearly laid out at
>
> http://www.stata.com/support/faqs/res/statalist.html#noanswer
>
> and stipulates _one repost only_.
>
> If you don't get an answer to this, sorry, but that's it. No one knows or wants to say.
>
> Nick
> n.j.cox@durham.ac.uk
>
> Allen Young
>
> I want to use the estimates obtained from fmm command to calculate the
> posterior probability of being in each of the components (latent
> classes). Please help
>
> *
> *   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/

--
Partha Deb
Professor of Economics
Director of Graduate Studies
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/