The model which is applicable to the longitudinal data in which the dropout process might give rise to the informative lost values. Specifically if the study protocol specifies the common set of n measurement times for all the subjects, and d is used to signify the subject's dropout time, with D=d if the values consequent to times d, d+1; ... ; n are missing and D=n+1 indicating that the subject did not drop out, then the statistical model includes the joint distribution of the observations y and d. This joint distribution can be given in two equivalent ways which are as follows,

 

the conditional probability