A note on some model selection criteria
From MaRDI portal
Data-adaptive priors are used as a descriptive tool to unify some of the previously proposed model selection criteria in a direct manner. An empirical Bayes criterion is constructed which penalizes for overfitting of model form as well as dimensionality.
Recommendations
Cites work
- A comparison of the information and posterior probability criteria for model selection
- A universal prior for integers and estimation by minimum description length
- An objective use of Bayesian models
- Estimating the dimension of a model
- Estimation of structure by minimum description length
- scientific article; zbMATH DE number 3675161 (Why is no real title available?)
- scientific article; zbMATH DE number 3591259 (Why is no real title available?)
- scientific article; zbMATH DE number 3635352 (Why is no real title available?)
- scientific article; zbMATH DE number 3444596 (Why is no real title available?)
Cited in
(3)
This page was built for publication: A note on some model selection criteria
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1084792)