Presented by: Bertrand S. Clarke, B.Sc., PhD, Professor and Chair, Department of Statistics, University of Nebraska-Lincoln

About the presentation: We compare the performance of six model average predictors – Mallows’ model averaging, stacking, Bayes model averaging, bagging, random forests, and boosting — to the components used to form them. In all six cases we provide conditions under which the model average predictor performs as well or better than any of its components. This is well known empirically, especially for complex problems, although few theoretical results seem to be available. Moreover, all six of the model averages can be regarded as Bayesian.  Mallows is asymptotically equivalent to Bayesian linear regression. Stacking is the Bayes optimal action in an asymptotic sense under several loss functions. Bayes model averaging is known to be the Bayes action under squared error. We show that bagging can be regarded as a special case of Bayes model averaging in an asymptotic sense. Random forests are a special case of bagging and hence likewise asymptotically Bayes. Boosted regression is a limit of Bayes optimal boosting classifiers.

To attend via Zoom, visit: https://pshealth.zoom.us/j/94860982030

To attend via the phone, dial 929-205-6099

The meeting ID is 948 6098 2030 and the passcode is 320367