Improving the estimation of Kendall's tau when censoring affects only one of the variables
Computational Statistics & Data Analysis
Permutation procedures with censored data
Computational Statistics & Data Analysis
A goodness-of-fit test for Archimedean copula models in the presence of right censoring
Computational Statistics & Data Analysis
| Hi-index | 0.03 |
This paper considers the nonparametric estimation of Kendall's tau for bivariate censored data. Under censoring, there have been some papers discussing the nonparametric estimation of Kendall's tau, such as Wang and Wells (2000), Oakes (2008) and Lakhal et al. (2009). In this article, we consider an alternative approach to estimate Kendall's tau. The main idea is to replace a censored event-time by a proper imputation. Thus, it induces three estimators, say @t@?"m"e"d"i"a"n, @t@?"m"e"a"n, and @t@?"m"o"d"e. We also apply the bootstrap method to estimate the variance of @t@?"m"e"d"i"a"n, @t@?"m"e"a"n and @t@?"m"o"d"e and to construct the corresponding confidence interval. Furthermore, we analyze two data sets by the suggested approach, and compare these practical estimators of Kendall's tau in simulation studies.