Frisch on testing of business cycle theories

An important identifying assumption for business cycle models is contained in the mathematical form of the model, which determines the nature of its possible movements. Tinbergen's and Frisch's original understanding of business cycle theories was that of a closed model, containing only endogenous variables. Differential equation models possessed the necessary features for identifying such a theory. However, in Tinbergen's later work for the League of Nations, the models consisted only of difference equations. Frisch showed that the close relationship between the observed movement and the mathematical model was lost in the transformation from differential to difference equations, for such models could only be identified in the presence of exogenous variables.