The use of components’ weights improves the diagnostic accuracy of a health-related index

Fragkiskos G. Bersimis, Dimosthenis Panagiotakos and Malvina Vamvakari

Communications in Statistics - Theory and Methods, 2019, vol. 48, issue 1, 141-164

Abstract: This work aims to evaluate whether the use of specific weights in each component of a health-related index, is associated with its diagnostic accuracy. In addition, the impact of a composite health-related index's components multitude is examined in relation to its classification ability. An un-weighted and various weighted indices were constructed using different weighting methods. The indices’ diagnostic ability was evaluated by using true positive rate, true negative rate, true rate, positive predictive value, negative predictive value and the area under the receiver operating characteristic curve. Weights used in this study were obtained from linear discriminant analysis and binary logistic regression. These indices were applied in both simulated and actual data; and a variety of scenarios was applied based on the distribution's parameters of the component variables and on the number of components used. Results indicate that weighted indices’ evaluation measures were significantly higher compared to the un-weighted one; whereas area under receiver operating characteristic curve was positively associated with the number of components of each index that were correlated with the outcome. Weighting of index's components, as well as greater number of components related to the investigated outcome should be recommended for the construction of accurate indices.

Date: 2019
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/03610926.2017.1388401 (text/html)
Access to full text is restricted to subscribers.

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:taf:lstaxx:v:48:y:2019:i:1:p:141-164

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/lsta20

DOI: 10.1080/03610926.2017.1388401

Access Statistics for this article

Communications in Statistics - Theory and Methods is currently edited by Debbie Iscoe

More articles in Communications in Statistics - Theory and Methods from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().