TIME-VARYING MULTIPRODUCT HEDGE RATIO ESTIMATION IN THE SOYBEAN COMPLEX: A SIMPLIFIED APPROACH

In developing optimal hedge ratios for the soybean processing margin, many authors have illustrated the importance of considering the interactions between the cash and futures prices for soybeans, soybean oil, and soybean meal. Conditional as well as time-varying hedge ratios have been examined, but in the case of multiproduct time-varying hedge ratios, the difficulty in estimation has been found to often outweigh any improvement in hedging effectiveness. This research examines the hedging effectiveness of the Risk Metrics procedure for estimating a time-varying covariance matrix for developing optimal hedge ratios for the soybean processing margin. The Risk Metrics method allows for a time-varying covariance matrix while being considerably easier to implement than multivariate GARCH (MGARCH) procedures. The Risk Metrics procedure has been advocated for use in developing Value-at-Risk estimates. While it provided considerable out-of-sample improvement in hedging effectiveness relative to a constant correlation MGARCH procedure, the Risk Metrics method provided only minimal improvement over a naive (1-to-1) hedging strategy. However, this research does illustrate the potential for the Risk Metrics methodology as a viable alternative to MGARCH procedures in a multiproduct hedging context.