A Theory Of Data-Oriented Identification With A Svar Application

I propose a method identification of structural vector autoregressions (SVARs) and simultaneous equations models (SEMs) with orthogonal structural shocks using testable identification restrictions. If some sparsity conditions are satisfied, the method produces a set of testable inclusions and exclusions, sufficient for the full identification. The method stems from the theory of probabilistic graphical models and from the theory of identification of SVARs and SEMs, merging them into a unified approach. In the application example, I estimate a SVAR monetary model of the US economy with 6 variables, where all but one identifying restrictions are testable. The method produces relatively narrow confidence intervals for the impulse-response functions, does not generate any anomalies such as the price puzzle, and reveals importance of informational channels through which news about structural shocks spread throughout the economy.