Bilgi Içeren Durdurma Varlığında Yinelemeli Olay Süreci
xmlui.mirage2.itemSummaryView.MetaDataShow full item record
Informative censoring is an important issue which have to be taken into account while modelling recurrent event data. Informative censoring might happen as a result of terminal event such as death or drop out of individuals in the study. Therefore, there is correlation between recurrent event time and terminal event time. In this study, we focused on the situation in which informative censoring happen as a result of death and modelled recurrent event data using homogeneous Poisson Process. We proposed two different models based on intensity functions and constructed the structure of correlations between recurrent event time and death time via shared frailty. Maximum likelihood estimates of parameters are obtained using EM and Metropolis-Hastings algorithms. The proposed models are applied on the well known bladder tumor study by Byar (1976). To show the validity of our proposed models, the simulation study is conducted for different scenarious.