Economics Research Institute Study Paper
Utah State University Department of Economics
Utilizing Cattle-Fax data, Granger-causality error correction models were used to determine whether the sufficient condition for market integration was satisfied for six cattle classes and twelve markets for which the necessary condition was already indicated. The bivariate results indicated that all the markets satisfied the sufficient condition, at least in one direction. The lightweight animals exhibited bidirectional causality more often than the heavy weight animals. Overall, bidirectional causality was most frequent for 400-pound steers and least frequent for 800-pound steers. The multivariate results were less optimal due to the restricted number of lags.
Muwanga, Gertrude S. and Snyder, Donald L., "Error Correction Models and Spatial Market Integration: Evidence for Cattle Markets in the U.S." (1997). Economic Research Institute Study Papers. Paper 131.