Efficient leave-one-out cross-validation for Bayesian non-factorized normal and Student-t models
Open Access
- 20 November 2020
- journal article
- research article
- Published by Springer Nature in Computational Statistics
- Vol. 36 (2) , 1243-1261
Abstract
Cross-validation can be used to measure a model’s predictive accuracy for the purpose of model comparison, averaging, or selection. Standard leave-one-out cross-validation (LOO-CV) requires that the observation model can be factorized into simple terms, but a lot of important models in temporal and spatial statistics do not have this property or are inefficient or unstable when forced into a factorized form. We derive how to efficiently compute and validate both exact and approximate LOO-CV for any Bayesian non-factorized model with a multivariate normal or Student- $$t$$ distribution on the outcome values. We demonstrate the method using lagged simultaneously autoregressive (SAR) models as a case study.
Keywords
All Related Versions
Funding Information
- Research Council of Finland (298742)
- Acedemy of Finland (313122)
- Technology Industries of Finland Centennial Foundation (70007503)
This publication has 27 references indexed in Scilit:
- A survey of Bayesian predictive methods for model assessment, selection and comparisonStatistics Surveys, 2012
- Predictive likelihood for Bayesian model selection and averagingInternational Journal of Forecasting, 2009
- Gaussian Processes in Machine Learning2004
- Proper multivariate conditional autoregressive models for spatial data analysisBiostatistics, 2003
- Bayesian model averaging: a tutorial (with comments by M. Clyde, David Draper and E. I. George, and a rejoinder by the authorsStatistical Science, 1999
- Multivariate Student-t regression models: Pitfalls and inferenceBiometrika, 1999
- STATISTICS FOR SPATIAL DATATerra Nova, 1992
- A Predictive Approach to Model SelectionJournal of the American Statistical Association, 1979
- An Inverse Matrix Adjustment Arising in Discriminant AnalysisThe Annals of Mathematical Statistics, 1951
- Adjustment of an Inverse Matrix Corresponding to a Change in One Element of a Given MatrixThe Annals of Mathematical Statistics, 1950