Modelling receiver operating characteristic curves using Gaussian mixtures

Amay S.M. Cheam and Paul D. McNicholas

Computational Statistics & Data Analysis, 2016, vol. 93, issue C, 192-208

Abstract: The receiver operating characteristic (ROC) curve is widely applied in measuring the performance of diagnostic tests. Many direct and indirect approaches have been proposed for modelling the ROC curve and, because of its tractability, the Gaussian distribution has typically been used to model both diseased and non-diseased populations. Using a Gaussian mixture model leads to a more flexible approach that better accounts for atypical data. The Monte Carlo method can be used to circumvent the absence of a closed-form for a functional form of the ROC curve. The proposed method, in which a Gaussian mixture is used in conjunction with the Monte Carlo method, performs favourably when compared to the crude binormal curve and the semi-parametric frequentist binormal ROC using the well-known LABROC procedure.

Keywords: Binormal curve; EM algorithm; Gaussian mixture distributions; LABROC; Mixture models; Monte Carlo method; ROC curve (search for similar items in EconPapers)
Date: 2016
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0167947315001073
Full text for ScienceDirect subscribers only.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:csdana:v:93:y:2016:i:c:p:192-208

DOI: 10.1016/j.csda.2015.04.010

Access Statistics for this article

Computational Statistics & Data Analysis is currently edited by S.P. Azen

More articles in Computational Statistics & Data Analysis from Elsevier
Bibliographic data for series maintained by Catherine Liu ().