109
reads

Rei-Yang Wang
2010-05-14  12:30 - 14:30
Room 405, Mathematics Research Center Building (ori. New Math. Bldg.)

This paper propose a criterion for model selection in prediction problems. As a contrast to AIC, the covariance inflation criterion reflects the model selection. It adjusts the training error by the average covariance of the predictions and responses, when the prediction rule is applied to permuted versions of the data set. This criterion can be applied to general prediction problems. They also obtain a measure of the effective number of parameters used by an adaptive procedure. And there will be comparison with other model selection procedures for orthogonal linear regression and others.