摘要

Survival analysis is a dynamic area in Statistics with many new methods in response to the practical needs in various applications. Applications of these methods have been widened from their historical use in cancer and reliability research to business, criminology, epidemiology, social and behavioral sciences. An important part of survival data analysis is to modeling or fitting of distribution. Once an appropriate statistical model for survival time has been constructed and its parameters estimated, its information can help predict survival, develop optimal treatment regimens, plan future clinical or laboratory studies and so on. In estimation of parameters of the functions, the nonparametric or distribution-free methods are quite easy to understand and apply. But they are less efficient than parametric methods when survival times follow a theoretical distribution and more efficient when no suitable theoretical distribution are known. In this paper, an attempt has been made to give an overview on the applications of most commonly used distributions by the researchers in survival analysis of cancer patient's viz. Exponential, Weibull, Lognormal, Gamma and Gompertz with their probability density function, survivorship function for survival function and hazard function.

  • 出版日期2010-12