We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Methodology

Title: Statistical inference for the two-sample problem under likelihood ratio ordering, with application to the ROC curve estimation

Abstract: Likelihood ratio ordering has been identified as a reasonable assumption in the two-sample problem in many practical scenarios. With this assumption, statisticians have proposed various methods in the estimation of the distributions of subpopulations, which consequently benefit the downstream inferences, such as the ROC curve and the associated summary statistic estimation. In this paper, under the likelihood ratio ordering assumption, we first propose a Bernstein polynomial method to model the distributions of both samples; we then estimate the distributions by the maximum empirical likelihood principle. The ROC curve estimate and the associated summary statistics are obtained subsequently. We compare the performance of our method with existing methods by extensive simulation studies. The application of our method is illustrated by a real-data example.
Comments: 18 pages, 1 figure
Subjects: Methodology (stat.ME)
Cite as: arXiv:2205.00505 [stat.ME]
  (or arXiv:2205.00505v1 [stat.ME] for this version)

Submission history

From: Pengfei Li [view email]
[v1] Sun, 1 May 2022 16:26:48 GMT (111kb,D)

Link back to: arXiv, form interface, contact.