Asymptotic normality and valid inference for Gaussian variational approximation

Publisher:
Institute of Mathematical Statistics
Publication Type:
Journal Article
Citation:
Annals of Statistics, 2011, 39 (1), pp. 2502 - 2532
Issue Date:
2011-01
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2010007039.pdf430.93 kB
Adobe PDF
We derive the precise asymptotic distributional behavior of Gaussian variational approximate estimators of the parameters in a single-predictor Poisson mixed model. These results are the deepest yet obtained concerning the statistical properties of a variational approximation method. Moreover, they give rise to asymptotically valid statistical inference. A simulation study demonstrates that Gaussian variational approximate confidence intervals possess good to excellent coverage properties, and have a similar precision to their exact likelihood counterparts.
Please use this identifier to cite or link to this item: