Smoothed Semi-parametric Regression in Heavy Censoring
Year: 2005
Type: article
Abstract: Regression with censored data is important in the inferences of the survival or failure time data. Present study focuses on the problem of modeling heavy censored survival data. In the class of Accelerated Failure Models, a new semi-parametric model is proposed to consider situations where the functional form of the effect of one or more covariates is unknown. This proposal is based on a model that does not need to assume any distribution for the survival time, or the proportionality of the hazard functions for different individuals. The effect of covariate is modeled directly on time, instead of the conditional probability or hazard. It is tried to improve estimation procedure to make inferences on parameters through the sum of the weighted squared residuals using the non-censored rate as proposed weights. A simulation study has been carried out to show the performance of the proposed estimation process and to analyze the effect of censorship.
Authors: Bahrawar Jan, S. W. Shah, Shurat Shah, Alamgir
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W2613050929
