Developing a goodness of fit test for a joint model of clustered survival and count data

Abstract

Mixed responses within clusters are becoming an increasingly popular phenomenon. However, this scenario is a complex one. This research is based on survival and count response data which occur frequently in medical studies. In the literature, there is no satisfactory goodness of fit test for a joint model of these two responses within a cluster setting. Therefore, the main objective of this study is to develop a GOF test for this scenario. Simulation studies were used to examine the type I error rate and the power of the test. An example was used to display the practicality of the test.

Keywords:

• Count
• Hierarchical data structure
• Joint model
• Mixed responses
• Simulation
• Survival