A Non-Random Dropout Model for Analyzing Longitudinal Skew-Normal Response (original) (raw)

Multivariate skew-normal distribution is employed for analyzing an outcome based dropout model for repeated measurements with non-random dropout in skew regression data sets. A probit regression is considered as the conditional probability of an observation to be missing given outcomes. A simulation study of using the proposed methodology and comparing it with a semi-parametric method, GEE, is provided. The standardized bias is used for comparison of different approaches. Furthermore, for investigation of efficiency of the methodology two applications are analyzed where observed information matrix is used to find the variances of the parameter estimates. In one of the applications a sensitivity analysis is also performed to investigate the change on the response model’s parameter estimates due to perturbation of drop-out model’s parameter of interest.