AN INFLUENCE APPROACH FOR SENSITIVITY ANALYSIS OF NON-RANDOM DROPOUT BASED ON THE COVARIANCE STRUCTURE

Abstract

A generalized Heckman model is used for the joint modeling of longitudinal continuousresponses and dropout in order to see the influence of a small perturbation of the elements of thecovariance structure on displacement of the likelihood. The perturbation from random dropout in thedirection of informative dropout is considered for Mastitis data.