Journal article
Binormal association-marginal models for empirically evaluating and comparing diagnostics
Statistical modelling, Vol.1(1), pp.49-64
04/01/2001
DOI: 10.1177/1471082X0100100105
Abstract
A new method for empirically evaluating and comparing two diagnostics is introduced. Specifically, correlated ordinal rating data from a paired-comparison study are modelled using a flexible, new class of binormal association-marginal (BAM) models. Among other things, these models, which are fitted via maximum likelihood (ML), afford efficient estimators of (i) the diagnostics' receiver operating characteristic curves and (ii) the level of manifest agreement between the diagnostics. BAM models use the latent binormal structure of classic signal detection theory to model each ordinal response marginal distribution. In contrast to bivariate binormal models, BAM models do not impose the added restriction that the ordinal responses have joint distributions that are determined by latent bivariate normal distributions. Instead, the association structure of the ordinal variables is directly specified using standard loglinear models. An ML fitting algorithm, which is related to those algorithms used to fit composite-link generalized linear marginal models, is introduced. The method is illustrated through the analyses of a neonatal radiograph data set and a simulated data set.
Details
- Title: Subtitle
- Binormal association-marginal models for empirically evaluating and comparing diagnostics
- Creators
- Joseph B Lang - University of IowaThor Aspelund - University of Iowa
- Resource Type
- Journal article
- Publication Details
- Statistical modelling, Vol.1(1), pp.49-64
- Publisher
- SAGE PUBLICATIONS LTD
- DOI
- 10.1177/1471082X0100100105
- ISSN
- 1471-082X
- eISSN
- 1477-0342
- Number of pages
- 16
- Grant note
- RO1 CA 62362 / National Institutes of Health
- Language
- English
- Date published
- 04/01/2001
- Academic Unit
- Statistics and Actuarial Science; Biostatistics
- Record Identifier
- 9984257624602771
Metrics
3 Record Views