Bayes Signification Tests in Linear Regression and Economic Applications

In this paper we will use the statistical inference, in particular the Bayes tests for the signification of the parameters of linear regression. We will consider the cases of known variance, and unknown variance respectively. Because we replace the Student tests by the Z tests in practice if the involved number of degrees of freedom is at least 30 , we can replace the case of unknown variance with that of known variance in our paper, if the above number of degrees of freedom is at least 30 . Because in economics we generally know the signs of the involved parameters, in practice the one-sided parametric tests are used. However, in this paper we will refer also to the two-sided Bayes test.