NWO Vici Award : Fundamentals of non-parametric Bayesian statistics

Bayesian statistics is being used on an increasingly larger scale in a range of complex statistical problems with very large numbers of unknown factors. At present we still have little or no fundamental understanding of the possibilities and limitations of this type of statistical procedure. In this project a theoretical framework will be developed that will make it possible to design optimal statistical procedures and to distinguish reliable procedures from unreliable ones.