Seminar Notice
Statistical Laboratory
Iowa State University
DATE AND TIME: Monday, April 11, 2005, 4:10 p.m.
PLACE: 319 Snedecor
SPEAKER: Xiaotong Shen, School of Statistics, University of Minnesota, St.
Paul
TITLE: Model Assessment, Selection and Averaging
ABSTRACT
In this talk, I will discuss a number of issues involved in model
assessment, selection and averaging from a prediction viewpoint. A general
technique of model assessment will be presented based on data
perturbation, yielding optimal selection, in particular, model selection and
combination. From a frequentist perspective, model combination over a
selected subset of modeling procedures is attractive, as it controls bias while
reducing variability, and hence yielding better performance in terms of the
accuracy of estimation and prediction. To realize the potential of model
combination, I will present methodologies for estimating the optimal tuning
parameter such as weights as well as subsets for combining via data
perturbation. Numerical examples are presented to illustrate main
aspects. This work is joint with Hsin-Cheng Huang.
COFFEE: 3:45 p.m., 104 Snedecor Hall
Jeanette La Grange
Department of Statistics
102 Snedecor
Iowa State University
Ames, IA 50010-1210
515 294-3440 (office)
515 294-4040 (fax)