Risk of Bayesian inference in misspecified models, and the sandwich covariance matrix (Q2869956)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Risk of Bayesian inference in misspecified models, and the sandwich covariance matrix |
scientific article; zbMATH DE number 6243277
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Risk of Bayesian inference in misspecified models, and the sandwich covariance matrix |
scientific article; zbMATH DE number 6243277 |
Statements
7 January 2014
0 references
posterior variance
0 references
quasi-likelihood
0 references
pseudo-true parameter value
0 references
interval estimation
0 references
Risk of Bayesian inference in misspecified models, and the sandwich covariance matrix (English)
0 references
This is a formal study of the model misspecification impact on the quality of Bayesian inference. The author suggests a superior mode of inference based on the ``sandwich'' covariance matrix. Finally, the proposed asymptotic decision analysis is compared to the standard Bayesian inference based on the likelihood model.
0 references
0.783544659614563
0 references
0.7566297054290771
0 references
0.7291611433029175
0 references
0.7238613367080688
0 references
