Maximum likelihood estimation of bivariate survival probabilities and its existence and uniqueness |
| |
Authors: | Shin-Soo Kang Kenneth J. Koehler |
| |
Affiliation: | 1. Department of Computer Science and Statistics , Kwan Dong University , Kangeon-Do, 215-800, South Korea;2. Department of Statistics , Iowa State University , Ames, Lowa, 50011-2020, USA |
| |
Abstract: | Nonparametric maximum likelihood estimation of bivariate survival probabilities is developed for interval censored survival data. We restrict our attention to the situation where response times within pairs are not distinguishable, and the univariate survival distribution is the same for any individual within any pair. Campbell's (1981) model is modified to incorporate this restriction. Existence and uniqueness of maximum likelihood estimators are discussed. This methodology is illustrated with a bivariate life table analysis of an angioplasty study where each patient undergoes two procedures. |
| |
Keywords: | Bivariate survival analysis Interval censoring Product limit estimator Greenwoods formula |
|
|