What is the deviance information criterion?
Add to your list(s)
Download to your calendar using vCal
If you have a question about this talk, please contact Nikolaos Demiris.
The deviance information criterion (DIC) is a Bayesian analogue of the Akaike Information Criterion (AIC). Although DIC is widely used as a model choice tool in Bayesian data analysis, its theoretical foundations
remain a subject of controversy. I will describe a theoretical derivation of DIC as a penalized loss function, where the penalty represents a rational price that must be paid for using the data twice:
once for parameter estimation, and again for model criticism. This derivation suggests some limitations on situations where DIC may be used, as well as providing improved asymptotic approximations and useful extensions.
This talk is part of the MRC Biostatistics Unit Seminars series.
This talk is included in these lists:
Note that ex-directory lists are not shown.
|