I am trying to fit a logistic regression model with missing covariates
using the EM algorithm by the method of weights (Ibrahim, JASA :
1990). I am having problems understanding the terminology of covariate
distribution.

How can one compute the covariate distribution. I think this is the
same approach taken by Bayesians in problems with missingness in the
covariates.

Thanks a lot.


Vumani
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to