Model Specification Tests Based on Artificial Linear Regressions

In this paper we develop an extremely general procedure for performing a wide variety of model specification tests by running artificial linear regressions and then using conventional significance tests. In particular, this procedure allows us to develop non-nested hypothesis tests for any set of models which attempt to explain the same dependent variable(s), even when the error specifications of the various models are not the same. For example, it is straightforward to test linear regression models against loglinear ones. These procedures are illustrated by an empirical application, in which we estimate and test several competing models of personal savings behavior in Canada