Factor-based mixture models in R?

[Edward Rigdon]

     I have a class on “latent variable mixture models” (meaning factor-based models where data are characterized by multiple distributions) coming up, and I don’t know how to estimate such a model in R.  Last year, my course was a mixture (ahem) of Mplus and lavaan (on the factor side), and of course the Mplus implementation of mixture models is very nice.  But is there any tool in R that estimates such models?  I don’t see the capability in lavaan yet, and my early searches have not turned up any other good option in R.  I’ll go back to Mplus if I have to (call it a chance to show some parallels between lavaan and Mplus), but I’ve done everything with R so far this semester—shame to have to crawl back to commercial software now.  (No offense meant to the Mplus team: their software is phenomenally excellent.)

--Ed Rigdon

[Mike Cheung]

Hi, Ed.

OpenMx can do mixture modeling though the model specification is not as easy as that in Mplus. http://openmx.psyc.virginia.edu/docs/openmx/latest/GrowthMixtureModel_Path.html

Mike

--

---------------------------------------------------------------------
 Mike W.L. Cheung               Phone: (65) 6516-3702
 Department of Psychology       Fax:   (65) 6773-1843
 National University of Singapore
 http://courses.nus.edu.sg/course/psycwlm/internet/
---------------------------------------------------------------------

On Sun, Nov 2, 2014 at 5:35 AM, Edward Rigdon <edward...@gmail.com> wrote:

     I have a class on “latent variable mixture models” (meaning factor-based models where data are characterized by multiple distributions) coming up, and I don’t know how to estimate such a model in R.  Last year, my course was a mixture (ahem) of Mplus and lavaan (on the factor side), and of course the Mplus implementation of mixture models is very nice.  But is there any tool in R that estimates such models?  I don’t see the capability in lavaan yet, and my early searches have not turned up any other good option in R.  I’ll go back to Mplus if I have to (call it a chance to show some parallels between lavaan and Mplus), but I’ve done everything with R so far this semester—shame to have to crawl back to commercial software now.  (No offense meant to the Mplus team: their software is phenomenally excellent.)

--Ed Rigdon

--
You received this message because you are subscribed to the Google Groups "lavaan" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lavaan+un...@googlegroups.com.
To post to this group, send email to lav...@googlegroups.com.
Visit this group at http://groups.google.com/group/lavaan.
For more options, visit https://groups.google.com/d/optout.

[Edward Rigdon]

Mike--

     Thanks!  Yes, model specification syntax is not as easy for OpenMx as it is for Mplus or lavaan--but maybe now is the time to play with Onyx, which can turn graphical model spec into syntax for OpenMx or Mplus.

--Ed Rigdon

Sent from my iPad