Case-deletion measures for models with incomplete data

This paper proposes several case-deletion measures for assessing the influe nce of an observation for complicated models with real missing data or hypo thetical missing data corresponding to latent random variables. The idea is to generalise Cook's (1977) approach to the conditional expectation of the complete-data loglikelihood function in the Em algorithm. On the basis of the diagnostic measures, a procedure is proposed for detecting influential observations. Two examples illustrate our methodology. We show that the met hod can be applied efficiently to a wide variety of complicated problems th at are difficult to handle by existing methods.