University of Cambridge > Talks.cam > Statistics > Default priors and model parametrization

  Browse A-Z

Log in

University Account External (via Google)

🔐 Log In

Information on

Subscribing to talksdetails Finding a talkdetails Adding a talkdetails Disseminating talksdetails Help and Documentationdetails

ℹī¸ Information

Default priors and model parametrization

Download to your calendar using vCal

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

This talk presents the development of classes of priors that ensure calibration of the resulting posterior inferences. These priors are built using asymptotic properties of likelihood inference and location model approximations to general models. The role of parameterization of the model in obtaining calibrated posterior inference for sub-parameters is described, and the proposed priors are related to Jeffreys’ prior and the Welch-Peers approach. Connections are made to so-called strong matching priors, which are data dependent priors derived by equating posterior marginal probabilities to conditional $p$-values, and the importance of targetting the prior on the parameter of interest.

This talk is part of the Statistics series.

This talk is included in these lists:

Note that ex-directory lists are not shown.

Other lists

Online, Zoom TBC Required lists for MLG Room SS03 (IN-PERSON ONLY)

Other talks

Online Causal Inference Seminar Overview of treatment for non-small cell lung cancer Necessary mechanisms for super AI and stopping hallucinations: The consistent reasoning paradox and the indeterminacy function Welcome Reception and Poster Session CodeScaler: Scaling Code LLM Training and Test-Time Inference via Execution-Free Reward Models Title and abstract tba
 

Š 2006-2025 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity