Title: Survival analysis with presence of informative censoring via nonparametric multiple imputation
Authors: Chengcheng Hu - University of Arizona (United States) [presenting]
Jeremy Taylor - University of Michigan (United States)
Chiu-Hsieh Hsu - University of Arizona (United States)
Abstract: A nonparametric multiple imputation approach is developed to recover information for censored observations while analyzing survival data with presence of informative censoring. A working shared frailty model is proposed to estimate the magnitude of informative censoring through estimating Kendall's tau, which is only used to determine the size of imputing risk set for each censored subject. Specifically, a larger tau indicates a smaller size of the imputing risk set. It has been shown that the posterior mean of frailty is a monotonic function of the observed time under the assumed models. Therefore, the observed times for subjects at risk are used to determine the imputing risk set for each censored subject. Simulation studies have shown that the nonparametric multiple imputation approach produces survival estimates close to the targeted values and coverage rates of the confidence intervals close to the nominal level, even in situations with a high degree of informative censoring.