In this article, an estimation method for length-biased and right censored data to assess the effects of risk factors under the under the semiparametric linear transformation model is proposed. It is observed that unlike an existing method by Shen et al. (Ref. 1) that is based on the ranks of observed failure times, the proposed new estimators are obtained from counting process-based unbiased estimating equations. In this connection, after giving basic notations, the estimating equations are constructed. Theoretical properties such as consistency and asymptotic normality for the estimators are derived under suitable regularity conditions. Simulations are conducted to study and evaluate the finite sample performance of the proposed method and the same is compared with that of the existing method. Further in order to illustrate the proposed method, real data set is also presented. (17 refs.)
展开▼
