Set of More Informative Models
- 82 Downloads
In this Chapter we survey that group of model selection procedures which have been derived within a decision framework in which there is no a priori information on the parameters. A loss function is explicitly assumed and the different criteria are based on the comparison of the estimated values of the corresponding risk function. Within these procedures we can distinguish two groups. The first one consists of those procedures that derive their risk functions assuming that one of the models is the true model. The second one includes those procedures for which the risk functions are obtained assuming in turn that each model is the true one.
KeywordsLoss Function True Model Risk Function Large Model Decision Framework
Unable to display preview. Download preview PDF.