利用差補方法分析半競爭風險資料之餘命迴歸模型

The Imputation Approach for Mean Residual Life Regression under Semi-Competing Risks Data

蔡尚儒(Shang-Ru Tasi)

指導教授 ： 謝進見

國立中正大學/理學院/數學系統計科學研究所/碩士(2015年)

並列摘要

This thesis focuses on the analysis of the mean residual life regression model under semi-competing risks data. Under semi-competing risks data, since the non-terminal event time is dependently censored by the terminal event time, we can not make inference on the non-terminal event time without extra assumptions. Thus, we use the Archimedean copula assumptions to specify the dependence between the non-terminal event time and the terminal event time. Under the Archimedean copula model assumption, we adopt the mean imputation method and the median imputation method to impute the non-terminal event time. Then, we apply the method by Magulurit and Zhang (1994) to estimate the regression coefficient. We examine the performance of the proposed approaches by simualtion studies. We also apply our suggested approachs to analyze the Bone Marrow Transplant data.

並列關鍵字

Semi-competing risks data ； Mean residual life regression model ； Archimedean Copula model

參考文獻

Clayton, D. G. (1978). A model for association in bivariate life tables and itsapplication in epidemiological studies of familial tendency in chronic diseaseincidence. Biometrika, 65: 141-151.

Fine, J., H. J. and Chappell, R. (2001). On semi-computing risks data. Biometrika, 88: 907-919.

Genest and Mackay (1986a). Copules archimediennes et familles de lois bidimensionnelles dont les marges sont donnees. The Canadian Journal of Statistics, 14: 145-159.

Genest and Mackay (1986b). The joy of copulas: Bivariate distributions with uniform marginals. The American Statistician, 40: 280-283.

Klein, J. and Moeschberger, M. (2003). Survival Analysis: Techniques for Censored and Truncated Data. New York: Springer.