Exact Inference for a New Flexible Hybrid Censoring Scheme
We introduce a new hybrid censoring scheme called general unified (progressive) hybrid censoring and study its properties by applying the modularization technique proposed in Górny and Cramer (Metrika 81(2):173–210, 2018a). For exponentially distributed lifetimes, we illustrate that already known progressive hybrid censoring models are included as particular cases. Further, we determine the exact distribution of the maximum likelihood estimators (MLEs) for an underlying exponential and uniform distribution under general unified progressive hybrid censoring and general unified hybrid censoring, respectively. The results are applied to construct exact confidence intervals.
KeywordsHybrid censoring Progressive censoring General unified hybrid censoring General unified progressive hybrid censoring Modularization Exponential distribution Uniform distribution B-spline Gamma distribution
We are grateful to an anonymous referee for the valuable comments and suggestions which led to both an extension and an improved presentation of the results.
- Childs A, Chandrasekar B, Balakrishnan N (2008) Exact likelihood inference for an exponential parameter under progressive hybrid censoring schemes. In: Vonta F, Nikulin M, Limnios N, Huber-Carol C (eds) Statistical models and methods for biomedical and technical systems, statistics for industry and technology. Birkhäuser, Boston, pp 323–334Google Scholar
- Cox DR, Oakes D (1985) Analysis of survival data. Monographs on statistics and applied probability, 2nd edn. Chapman and Hall, LondonGoogle Scholar
- Górny J (2017) A new approach to hybrid censoring. Ph.D. thesis, RWTH Aachen University, Aachen, Germany. https://doi.org/10.18154/RWTH-2017-05192