Presenter: Johannes Buchner, MPE
Title: Deciding between models (discussion)
Abstract: What exactly do we mean when we compare models with each other? Different styles of model comparison corresponding to different questions will be presented, including based on information content, predictive power and parsimony. Bayesian inference can deliver relative model probabilities and insightful visualisations, but does not make decisions. Tools from statistics and machine learning can quantify false decision rates for Bayes factor thresholds. This enables a frequentist assessment (how well does a procedure work) on a Bayesian method (what is the relative probability of two models), while also reducing prior influences.
Paper: Appendix C of https://arxiv.org/abs/1402.0004
As usual, the format will be a short presentation followed by plenty of discussion.
Please let all interested know about the Journal Club (get them to send their Email address). We are looking forward to lively discussions.
Topic: ODSL Forum
Time: Oct 8, 2021 02:00 PM Amsterdam, Berlin, Rome, Stockholm, Vienna
Join Zoom Meeting
Meeting ID: 632 9359 2746