A Bootstrap Method for Identifying and Evaluating a Structural Vector Autoregression

Graph-theoretic methods of causal search based in the ideas of Pearl (2000), Spirtes,Glymour, and Scheines (2000), and others have been applied by a number of researchersto economic data, particularly by Swanson and Granger (1997) to the problem of findinga data-based contemporaneous causal order for the structural autoregression (SVAR),rather than, as is typically done, assuming a weakly justified Choleski order. Demiralpand Hoover (2003) provided Monte Carlo evidence that such methods were effective,provided that signal strengths were sufficiently high. Unfortunately, in applications toactual data, such Monte Carlo simulations are of limited value, since the causal structureof the true data-generating process is necessarily unknown. In this paper, we present abootstrap procedure that can be applied to actual data (i.e., without knowledge of the truecausal structure). We show with an applied example and a simulation study that theprocedure is an effective tool for assessing our confidence in causal orders identified bygraph-theoretic search procedures.