B1355
Title: Accelerated functional failure time models with error-prone response and its application to cancer data
Authors: Li-Pang Chen - National Chengchi University (Taiwan) [presenting]
Abstract: As a specific application of survival analysis, one of the the main interests in medical studies is to analyse a specific cancer's lifetime. Typically, gene expressions are treated as covariates to characterize the survival time. In the framework of survival analysis, the accelerated failure time (AFT) model in the parametric form is perhaps a common approach. However, gene expressions are possibly non-linear, and the survival time as well as censoring status are subject to measurement error. The aim is to tackle those complex features simultaneously. It is first corrected for measurement error in survival time and censoring status and used to develop a corrected Buckley-James estimator. After that, the boosting algorithm is used with the cubic spline estimation method to recover the non-linear relationship between covariates and survival time iteratively. Theoretically, the validity of measurement error correction and estimation procedure is justified. Numerical studies show that the proposed method improves estimation performance and can capture informative covariates. The methodology is primarily used to analyze the breast cancer data provided by the Netherlands Cancer Institute for research.
