University of Cambridge > > Probability Theory and Statistics in High and Infinite Dimensions > Adaptive nonparametric credible balls

Adaptive nonparametric credible balls

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact clc32.

Credible sets are central sets in the support of a posterior probability distribution, of a prescribed posterior probability. They are widely used as means of uncertainty quantification in a Bayesian analysis. We investigate the frequentist coverage of such sets in a nonparametric Bayesian setup. We show by example that credible sets can be much too narrow and misleading, and next introduce a concept of `polished tail’ parameters for which credible sets are of the correct order. The latter concept can be seen as a generalisation of self-similar functions as considered in a recent paper by Giné. Joint work with Botond Szabó and Harry van Zanten.

This talk is part of the Probability Theory and Statistics in High and Infinite Dimensions series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2024, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity