Identification may be established either by the examination of the specification of the structural model, or by the examination of the reduced form of the model.

Traditionally identification has been approached via the reduced form. In the sobsequent section we will examine both approaches. However, the reduced form approach is conceptually confusing and computationally more difficult than the structural model approach, because it requires the derivation of the reduced form first and then examination of the values of the determinant formed from some of the reduced form coefficients. The structural form approach is simpler and more useful.

In this unit we will only discuss the structural form approach of identification. In applying the identification rules we should ignore the constamt term if it is present in each equations, or, if it is present in some of the equations then we have to retain it and we need to treat it like separate variable. In this case we must include in the set of varijkdes a dummy variable (say X, ), which would always take on the value 1.