"em" Object
DESCRIPTION:
Objects of class
em that result from using the EM
algorithm.
GENERATION:
This class of objects may be returned as the
algorithm component in a class
"missmodel" object.
METHODS:
The class
"em" has associated methods:
.
STRUCTURE:
The class
"em" object consists of a
list containing the following values:
VALUE:
- last
-
the number of final iteration parameter estimates to save.
Can be specified through the
control object in the call generating
the class
"em" object.
- likelihood
-
the log likelihood (or the log posterior density) at the final
parameter estimates.
- reldif
-
the maximum absolute relative change in a parameter estimate on the
final iteration.
- likdif
-
the difference in the log-likelihood (or the log posterior density) on
the final two iterations.
SEE ALSO:
,
,
,
,
,
,
,
,
.