Identifiability of masking probabilities in competing risks models with emphasis on Weibull models |
| |
Authors: | Ye Liang Dongchu Sun |
| |
Institution: | 1. Department of Statistics, Oklahoma State University, Stillwater, Oklahoma, USAye.liang@okstate.edu;3. Department of Statistics, University of Missouri, Columbia, Missouri, USA;4. School of Finance and Statistics, East China Normal University, Shanghai, China |
| |
Abstract: | AbstractLifetime data with masked failure causes arise in both reliability engineering and epidemiology. The phenomenon of masking occurs when a subject is exposed to multiple risks. A competing risks model with masking probabilities is widely used for the masked failure data. However, in many cases, the model suffers from an identification problem. We show that the identifiability of masking probabilities depends on both the structure of data and the cause-specific hazard functions. Motivated by this result, two existing solutions are reviewed and further improved. |
| |
Keywords: | Identifiability Lifetime data Masked failure causes Overparameterization Proportional hazard functions |
|
|