Model selection in linear regression

We consider the multiple regression model Y = X(0) beta + epsilon, where Y and epsilon are n-vector random variables, X(0) is an n x m design matrix and beta is an in-vector of unknown regression parameters. It is well known that different information theoretic criteria with proper choice of penalty function can be used to choose the correct model. In this paper we have done an extensive simulation study to choose the proper penalty function, by using different models and using different error random variables.