首页 | 本学科首页   官方微博 | 高级检索  
     检索      


Incorporating diagnostic accuracy into the estimation of discrete survival function
Authors:Abidemi K Adeniji  Steven H Belle  Abdus S Wahed
Institution:1. Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, Connecticut, CT, USA;2. Department of Epidemiology, University of Pittsburgh, 130 DeSoto Street Pittsburgh, PA 15261, USA;3. Department of Biostatistics, University of Pittsburgh, 130 DeSoto Street, Pittsburgh, PA 15261, USA
Abstract:Empirical distribution function (EDF) is a commonly used estimator of population cumulative distribution function. Survival function is estimated as the complement of EDF. However, clinical diagnosis of an event is often subjected to misclassification, by which the outcome is given with some uncertainty. In the presence of such errors, the true distribution of the time to first event is unknown. We develop a method to estimate the true survival distribution by incorporating negative predictive values and positive predictive values of the prediction process into a product-limit style construction. This will allow us to quantify the bias of the EDF estimates due to the presence of misclassified events in the observed data. We present an unbiased estimator of the true survival rates and its variance. Asymptotic properties of the proposed estimators are provided and these properties are examined through simulations. We evaluate our methods using data from the VIRAHEP-C study.
Keywords:product limit estimation  misclassification  measurement error  diagnostic testing  binary classification
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号