Sankhya: The Indian Journal of Statistics1997, Volume 59, Series A, Pt. 2, 242-267
MODELLING EXPONENTIAL SURVIVAL DATA WITH DEPENDENT CENSORING
AMIT BHATTACHARYA, SmithKline Beecham Pharmaceuticals, Harlow
SUMMARY. An absolutely continuous bivariate model is proposed for modeling survival data with random censoring where the censoring pattern (or scheme) and the failure pattern are not statistically independent. In this paper, the failure and the censoring distributions are considered to be exponential with different means. The identifiability of the model is established, and the asymptotic normality of the maximum likelihood estimator of the parameters is proved. A simulation study is performed in support of the asymptotic normality. A practical data set is considered for the purpose of fitting the model to the data.
AMS (1991) subject classification. 62N02, 62F12, 62N05.
Key words and phrases. Survival function, competing risks, Kaplan Meier estimator, dependent censoring, crude and net hazard, rho-function, identified minima, identifiability, maximum likelihood estimator, asymptotic normality.
Full paper (PDF)
This article in Mathematical Reviews