B1205
Title: Goodness of fit tests for partly interval censored survival data
Authors: Najmeh Nakhaeirad - University of Pretoria (South Africa) [presenting]
Vahid Fakoor - Department of Statistics Ferdowsi University of Mashhad (Iran)
Din Chen - University of Pretoria (South Africa)
Abstract: In clinical and epidemiological research, depending on the frequency of clinic visits, the failure times may be ascertained exactly or only known within certain intervals, which are referred to as partly interval-censored data. In practice, researchers frequently prefer to make specific assumptions about the underlying distribution of the failure times. To address this preference, the goodness of fit tests designed for handling partly interval-censored data is introduced, which has received little attention in the literature. The proposed tests are based on the Cramer von Mises type of test and a divergence measure. It is demonstrated that these tests are asymptotically consistent. The effectiveness of the proposed approaches for small samples is evaluated through Monte Carlo simulations, and a real dataset is also analyzed to illustrate the practical application of the proposed tests.
