Skip to content

Loading Events

« All Events

  • This event has passed.

Biostatistics seminar: “Predictive Bayes Factors”

March 5 at 12:00 pm - 1:00 pm

“Predictive Bayes Factors” will be presented by Shouhao Zhou, PhD, assistant professor in the Division of Biostatistics and Bioinformatics, Department of Public Health Sciences, Penn State College of Medicine, as part of the department’s biostatistics seminar series.

The Bayes factors approach has been widely used as an applied model selection tool by comparing posterior model probabilities. However, it has also been extensively misused because the posterior model probabilities calculated by Bayes factors are only for the original Bayesian models with parameter distributions fixed at the prior. If the candidate models of interest are fitted Bayesian models with parameter distributions updated to posterior by the observed data, traditional Bayes factors approach is not feasible. Therefore, Zhou’s group proposes a new approach, predictive Bayes factors, for predictive model selection of fitted Bayesian models via posterior model probability. It asymptotically estimates the out-of-sample log-likelihood ratio without taking the risk of using the data twice in both model estimation and validation. In practice, it dramatically reduces sensitivity to variations in the prior and totally avoids the Lindley’s paradox in testing point null hypothesis. The numerical example illustrates the discrepancy between the standard Bayes factors and predictive Bayes factors when the fitted models are of interest in practice.

In addition to the in-person presentation in the Academic Support Building on the College of Medicine campus in Hershey, participants may also join by calling 669-900-6833, meeting ID 322 043 373, or via Zoom webinar.

Light refreshments will be served to in-person attendees.



March 5
12:00 pm - 1:00 pm
Event Category:


Department of Public Health Sciences


Academic Support Building (ASB) Room 2200G
90 Hope Drive
Hershey, PA 17033 United States
+ Google Map