A Framework for the Evaluation of Statistical Prediction Models

Michael W Kattan; Thomas A Gerds

doi:10.1016/j.chest.2020.03.005

A Framework for the Evaluation of Statistical Prediction Models

Chest. 2020 Jul;158(1S):S29-S38. doi: 10.1016/j.chest.2020.03.005.

Authors

Michael W Kattan¹, Thomas A Gerds²

Affiliations

¹ Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH. Electronic address: [email protected].
² Department of Public Health, Section of Biostatistics, University of Copenhagen, Copenhagen, Denmark.

PMID: 32658649
DOI: 10.1016/j.chest.2020.03.005

Abstract

We have entered the era of "big data" and with that the explosion of interest in prediction modeling. With this explosion comes the challenge of evaluating statistical prediction models, both from the standpoint of an author as well as a reviewer. This article provides guidance for the evaluation and critique of a statistical prediction model. Hopefully, this will improve the quality of statistical prediction modeling studies and facilitate their review.

Keywords: medical decision-making; outcomes; prediction models.

Publication types

Review

MeSH terms

Evaluation Studies as Topic*
Forecasting / methods*
Humans
Models, Statistical*
Research Design / statistics & numerical data*