R help - Mar 2012 - FIML in R

Does anyone know if someone is developing full-information maximum likelihood
(FIML) estimation algorithms for basic regression functions, like glm()?  I
think that having FIML options for workhorse functions that typically use ML
would give R an edge over other statistical software, given how well FIML
performs in missing data situations compared to ML.

While my current level of programming ability isn't up to the task, I bring
this up to a) see if there is something already underway, and if not b) perhaps
spark interest in doing so.  I think much of the raw material is already out
there: there are other packages that use the EM algorithm, and other packages
that search out missing data patterns (e.g., md.pattern() in the mice package). 
To facilitate adoption, it might be useful to write a FIML module that other
package maintainers can incorporate into their code.

I'd be interested in updates on this, and in your thoughts on this more
generally.  Also, please let me know if there is a forum better suited for this
kind of discussion.

Andrew Miles

I completely agree that the development of full-information maximum
likelihood (FIML) estimation for use in packages like lm, lme, lmer, etc.
would make R much more attractive.  FIML is better than other approaches to
missing data (e.g., multiple imputation; see Graham, Olchowski, & Gilreath,
2007), and it is much better than what R currently uses for most functions
(listwise deletion).  Right now, the best implementation of FIML is in Mplus
(http://statmodel.com/), which is useful in the structural equation model
(SEM) framework.  There is an R package that incorporates FIML in SEM called
OpenMx (http://openmx.psyc.virginia.edu/openmx-features), in which one could
run multiple regression with FIML.  It would be nice to see FIML estimation
incorporated into other functions, as well though, as FIML seems to be the
way to go.  I second the call to get the R community talking about FIML.

Thank you Andrew for starting this -- it's long overdue.  Hopefully, other
people are interested, as well.

--
View this message in context:
http://r.789695.n4.nabble.com/FIML-in-R-tp4515074p4538436.html
Sent from the R help mailing list archive at Nabble.com.