A Dynamic Model of U.S. Sugar-Related Markets: A Cointegrated Vector Autoregression Approach

The methods of the cointegrated vector autoregression (VAR) model are applied to monthly U.S. markets for sugar and for sugar-using markets for confectionary, soft drink, and bakery products. Primarily a methods paper, we apply Johansen and Juselius' advanced procedures to these markets for perhaps the first time, with focus on achievement of a statistically adequate model through analysis of a battery of advanced statistical diagnostic tests and on exploitation of the system's cointegration properties through rank restrictions, statistically supported hypotheses test restrictions, and inference. The VEC model results illuminate the estimates of crucial policy-relevant market parameters that drive these markets, as well as the dynamic nature of the relationships linking these sugar-based markets.