The survival function estimation of current status data with dependent censoring

• Main Authors: Yung-Yu Chen, 陳永諭
• Other Authors: Jin-Jian Hsieh
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/zq8bd6

碩士 === 國立中正大學 === 數學系統計科學研究所 === 102 === This thesis focuses on the estimation of the survival function of the failure time under the current status data. Because the failure time may be correlated with the observation time in the practice, we would like to investigate the estimation of the survival function of the failure time under dependent censoring. We use the Archimedean Copula model to specify the dependency between the failure time and the observation time. Under the Archimedean Copula model assumption, we adopt a redistribution algorithm to estimate the survival function of the failure time. We examine the finite-sample performance of the proposed approach by simulation studies and compared it with a pool-adjacent-violators type algorithm (Titman, 2013). We also apply our proposed methodology to analyze a practical tumorigenicity data.