FISHER INFORMATION IN ORDERED RANDOMLY CENSORED DATA WITH APPLICATIONS TO CHARACTERIZATION PROBLEMS

The proportional hazards model of Koziol-Green is often considered in survival analysis. If the lifetime and censoring random variables are independent, the Koziol-Green model implies that the variable indicating whether the observa- tion is censored or not does not contain Fisher information about the parameters of the underlying lifetime distribution. The Koziol-Green model, however, is not uniquely characterized by this result on the lackof Fisher information in the censor- ing indicator. Given the ordered randomly censored lifetimes with corresponding indicators, we obtain a necessary and sufficient condition, weaker than the Koziol- Green model, which ensures that a set of any number of censoring indicators does not contain Fisher information about the parameters of the lifetime distribution. Under this weaker condition, the results are applied to characterize the Weibull distribution within the class of scale parameter families of lifetime distributions and the factorization of the hazard function in terms of the Fisher information in randomly censored data.

[1]  Bradley Efron,et al.  FISHER'S INFORMATION IN TERMS OF THE HAZARD RATE' , 1990 .

[2]  M. Hollander,et al.  Small-Sample Results for the Kaplan-Meier Estimator , 1982 .

[3]  Joseph L. Gastwirth,et al.  On the Fisher information in randomly censored data , 2001 .

[4]  Sangun Park,et al.  Fisher Information in Order Statistics , 1996 .

[5]  Z. A. Abo-Eleneen,et al.  Fisher Information in an Order Statistic and its Concomitant , 2002 .

[6]  M. Wells,et al.  Quantile Comparison Functions in Two-Sample Problems, with Application to Comparisons of Diagnostic Markers , 1996 .

[7]  Calyampudi Radhakrishna Rao,et al.  Linear Statistical Inference and its Applications , 1967 .

[8]  S. Chakraborti,et al.  An Overview of Precedence‐Type Tests for Censored Data , 1997 .

[9]  A. Kimball,et al.  Chemical protection against ionizing radiation. I. Sampling methods for screening compounds in radiation protection studies with mice. , 1957, Radiation research.

[10]  J. Gastwirth,et al.  WHERE IS THE FISHER INFORMATION IN AN ORDERED SAMPLE , 2000 .

[11]  G. Zheng.  A characterization of the factorization of hazard function by the Fisher information under Type II censoring with application to the Weibull family , 2001 .

[12]  Joseph L. Gastwirth,et al.  CONTROL PERCENTILE TEST PROCEDURES FOR CENSORED DATA , 1988 .

[13]  Winfried Stute.  Strong consistency under the Koziol—Green model , 1992 .

[14]  Lajos Horváth,et al.  On the Koziol-Green model for random censorship , 1981 .

[15]  James A. Koziol,et al.  A Cramér-von Mises statistic for randomly censored data , 1976 .

[16]  Richard A. Johnson,et al.  Exact fisher information for censored samples and the extended hazard rate functions , 1979 .

[17]  B. L. S. Prakasa Rao.  Remarks on Cramer-Rao type integral inequalities for randomly censored data , 1995 .

[18]  Pawlitschko Jörg.  A Comparison of Survival Function Estimators in the Koziol–Green Model , 1999 .

[19]  J. Gastwirth.  Asymptotically Most Powerful Rank Tests for the Two-Sample Problem with Censored Data , 1965 .