Limitations of “Limitations of Bayesian Leave-one-out Cross-Validation for Model Selection”

In an earlier article in this journal, Gronau and Wagenmakers ( 2018 ) discuss some problems with leave-one-out cross-validation (LOO) for Bayesian model selection. However, the variant of LOO that Gronau and Wagenmakers discuss is at odds with a long literature on how to use LOO well. In this discu...

Full description

Saved in:
Bibliographic Details
Published inComputational brain & behavior Vol. 2; no. 1; pp. 22 - 27
Main Authors Vehtari, Aki, Simpson, Daniel P., Yao, Yuling, Gelman, Andrew
Format Journal Article
LanguageEnglish
Published Cham Springer International Publishing 01.03.2019
Subjects
Behavioral Science and Psychology
Cognitive Psychology
Mathematical Models of Cognitive Processes and Neural Networks
Psychological Methods/Evaluation
Psychology
Statistical convenience
M-open
Reality
M-closed
Principle of complexity
Online AccessGet full text

Cover

More Information
Summary:In an earlier article in this journal, Gronau and Wagenmakers ( 2018 ) discuss some problems with leave-one-out cross-validation (LOO) for Bayesian model selection. However, the variant of LOO that Gronau and Wagenmakers discuss is at odds with a long literature on how to use LOO well. In this discussion, we discuss the use of LOO in practical data analysis, from the perspective that we need to abandon the idea that there is a device that will produce a single-number decision rule.
ISSN:2522-0861
2522-087X
DOI:10.1007/s42113-018-0020-6