Multistep estimation of regression coefficients in a linear model with censored survival data

摘要: This paper introduces a multi-step procedure for estimating the regression coefficients in linear model when dependent variable of interest is randomly right censored transform survival, i.e., log lifetime. The closely related to that introduced by Buckley and James (1979). Using large sample properties developed authors (1981a), asymptotic consistency normality are seen hold each iterate original estimator. A limited simulation study examines small behavior procedure.