Multiscale inference about a density

• 成果类型: Article
• 署名作者: Duembgen, Lutz; Walther, Guenther
• 署名单位: University of Bern; Stanford University
• 刊物名称: ANNALS OF STATISTICS
• ISSN/ISSBN: 0090-5364
• DOI: 10.1214/07-AOS521
• 发表日期: 2008
• 页码: 1758-1785
• 关键词: excess mass; normalized spacings; tests; multimodality; modality; contour

摘要

We introduce a multiscale test statistic based on local order statistics and spacings that provides simultaneous confidence statements for the existence and location of local increases and decreases of a density or a failure rate. The procedure provides guaranteed finite-sample significance levels, is easy to implement and possesses certain asymptotic optimality and adaptivity properties.