Full information maximum likelihood (FIML) uses only
the observed data. To over-simplify, FIML partitions
the cases into subsets with the same patterns of
missing observations. All available statistical
information is extracted from each subset and all
cases are retained in the analysis.
More formally, mean vectors and covariance matrices
are formed for cases that have the same pattern of
observed data. Once the mean vectors and covariance
matrices have been formed, the FIML approach of
Arbuckle (1996) uses the fact that for the i-th case,
the log-likelihood function can be expressed as:
log Li=Ci - 1/2log|Si| - 1/2(xi-mui)'S^-1(xi-mui)
and the log likelihood of the entire sample is the sum
of the individual log likelihoods. The likelihood is
maximized in terms of the parameters of the model.
As a practical matter, FIML can be easily implemented
in multiple regression and factor analysis models to
handle missing data to that no cases are lost.
Scott Millis
--- [email protected] wrote:
> Then the analysis would be only on the observed
> data? Or is there any hidden
> imputation type of process going on? Thanks!
Scott R Millis, PhD, MEd, ABPP (CN,CL,RP), CStat
Professor & Director of Research
Dept of Physical Medicine & Rehabilitation
Wayne State University School of Medicine
261 Mack Blvd
Detroit, MI 48201
/ `--'(
< [] []////////|:::)
\_.--.(
Email: [email protected]
Tel: 313-993-8085
Fax: 313-966-7682
*********************************************************
This electronic message may contain information that is confidential and/or legally privileged. It is intended only for the use of the individual(s) and entity named as recipients in the message. If you are not an intended recipient of this message, please notify the sender immediately and delete the material from any computer. Do not deliver, distribute or copy this message, and do not disclose its contents or
take any action in reliance on the information it contains. Thank you.
*
* 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/
