Bayesian likelihood methods for estimating the end point of a distribution

• 成果类型: Article
• 署名作者: Hall, P; Wang, JZ
• 署名单位: Western Sydney University; Australian National University
• 刊物名称: JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
• ISSN/ISSBN: 1369-7412
• DOI: 10.1111/j.1467-9868.2005.00523.x
• 发表日期: 2005
• 页码: 717-729
• 关键词: maximum-likelihood; truncated distribution; translation parameter; models

摘要

We consider maximum likelihood methods for estimating the end point of a distribution. The likelihood function is modified by a prior distribution that is imposed on the location parameter. The prior is explicit and meaningful, and has a general form that adapts itself to different settings. Results on convergence rates and limiting distributions are given. In particular, it is shown that the limiting distribution is non-normal in non-regular cases. Parametric bootstrap techniques are suggested for quantifying the accuracy of the estimator. We illustrate performance by applying the method to multiparameter Weibull and gamma distributions.