Representation theorems in Bayesian statistics

by József Zsolt Bernád (Forschungszentrum Jülich, Germany)

Monday 6 Jan 2025, 10:00 → 11:00 Europe/Budapest

Médiaterem (2. épület)

Throughout this talk, I will discuss several representation theorems about observable random quantities. In the beginning, the concept of exchangeability with some examples is discussed, which leads to representations that justify and clarify the use and interpretation of such familiar concepts as likelihoods and prior distributions. As we establish justification for using Bayesian methods, it is shown that beliefs that possess certain additional invariance properties lead to representations that include well-known specific forms of parametric distributions, such as normal and exponential distributions. Then, I will discuss partial exchangeability judgments about data structures. Finally, a brief overview of the differences between the objective and subjective probabilities will be presented.

Zoom meeting: https://cern.zoom.us/j/67590004845?pwd=TmhMaVpCOFhQZ3RCRUJwVUVMSFV0dz09