Semiparametric inference for a quantile comparison function with applications to receiver operating characteristic curves

• 成果类型: Article
• 署名作者: Li, G; Tiwari, RC; Wells, MT
• 署名单位: University of California System; University of California Los Angeles; University of North Carolina; University of North Carolina Charlotte; Cornell University
• 刊物名称: BIOMETRIKA
• ISSN/ISSBN: 0006-3444
• DOI: 10.1093/biomet/86.3.487
• 发表日期: 1999
• 页码: 487502
• 关键词: of-fit tests; diagnostic markers; WEAK-CONVERGENCE; censored-data; ROC curve; estimator; bootstrap; models; FAMILY; hazard

摘要

In studies to compare two samples, more information may be available on one treatment than the other. When one population is modelled parametrically and the other nonparametrically, we study large sample properties of a semiparametric sample quantile comparison function and show that it can have substantially smaller asymptotic variance than its nonparametric counterpart, especially near the boundaries. We describe applications to both two-sample tests and receiver operating characteristic curves.