Global hypothesis test to compare the likelihood ratios of multiple binary diagnostic tests with ignorable missing data

Marín-Jiménez, Eugenia; Roldán Nofuentes, José Antonio

MarinJimenez_MissingData.pdf (146.9Kb)

Identificadores

URI: http://hdl.handle.net/10481/34829

ISSN: 1696-2281

ISSN: 2013-8830

Exportar

Editorial

Institut d'Estadística de Catalunya (Idescat)

Materia

Global hipothesis test

Partial verification

Positive and negative likelihood ratio

Date

2014

Referencia bibliográfica

Marín-Jiménez, E.; Roldán-Nofuentes, J.A. Global hypothesis test to compare the likelihood ratios of multiple binary diagnostic tests with ignorable missing data. Sort, 38(2): 305-324 (2014). [http://hdl.handle.net/10481/34829]

Sponsorship

This research was supported by the Spanish Ministry of Science, Grant Number MTM 2012-35591.

Abstract

In this article, a global hypothesis test is studied to simultaneously compare the likelihood ratios of multiple binary diagnostic tests when in the presence of partial disease verification the missing data mechanism is ignorable. The hypothesis test is based on the chi-squared distribution. Simulation experiments were carried out to study the type I error and the power of the global hypothesis test when comparing the likelihood ratios of two and three diagnostic tests respectively. The results obtained were applied to the diagnosis of coronary stenosis.

Collections

DEIO - Artículos