Bayesian hierarchical stacking: some models are (somewhere) useful
From MaRDI portal
Abstract: Stacking is a widely used model averaging technique that asymptotically yields optimal predictions among linear averages. We show that stacking is most effective when model predictive performance is heterogeneous in inputs, and we can further improve the stacked mixture with a hierarchical model. We generalize stacking to Bayesian hierarchical stacking. The model weights are varying as a function of data, partially-pooled, and inferred using Bayesian inference. We further incorporate discrete and continuous inputs, other structured priors, and time series and longitudinal data. To verify the performance gain of the proposed method, we derive theory bounds, and demonstrate on several applied problems.
Cites work
- 10.1162/153244304773936090
- A Bayes interpretation of stacking for \(\mathcal{M}\)-complete and \(\mathcal{M}\)-open settings
- A survey of Bayesian predictive methods for model assessment, selection and comparison
- Approximate leave-future-out cross-validation for Bayesian time series models
- Bayesian model averaging: A tutorial. (with comments and a rejoinder).
- Combining Estiamates in Regression and Classification
- Comparison of Bayesian predictive methods for model selection
- Covariate shift adaptation by importance weighted cross validation
- Data augmentation for support vector machines
- Dynamic Bayesian Forecasting of Presidential Elections in the States
- scientific article; zbMATH DE number 578421 (Why is no real title available?)
- scientific article; zbMATH DE number 1522714 (Why is no real title available?)
- scientific article; zbMATH DE number 7049770 (Why is no real title available?)
- Improving predictive inference under covariate shift by weighting the log-likelihood function
- Input-dependent estimation of generalization error under covariate shift
- Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC
- Seeking efficient data augmentation schemes via conditional and marginal augmentation
- Sparse ensembles using weighted combination methods based on linear programming
- Stacked regressions
- Statistical modeling: The two cultures. (With comments and a rejoinder).
- Structured priors for sparse probability vectors with application to model selection in Markov chains
- Using stacking to average Bayesian predictive distributions (with discussion)
Cited in
(5)
This page was built for publication: Bayesian hierarchical stacking: some models are (somewhere) useful
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6203242)
