ABSTRACT

This chapter is concerned with an important class of problems encountered in the estimation of economy-wide econometric models. It considers the choice of an eligible list of instruments, the criterion for eligibility being near zero correlation in the probability limit with the disturbance term from the given equation. The chapter examines how one should go about selecting from it the instrumental variables that will actually be employed. It discusses that a good instrumental variable should directly or indirectly causally influence the variables in the equation to be estimated in a way independent of the other instrumental variables, and that the more direct such influence is, the better. In general an instrumental variable should be known to cause the included variables in the equation, at least indirectly. The procedure gives an a priori preference order on the set of instrumental variables relative to a given zero causal order endogenous variable.