Tests of the Difference between In-Sample and Post-Sample Hedging Effectiveness

Hedging effectiveness is proportional price-risk reduction achieved by hedging. Typically, hedging studies estimate hedging effectiveness for the sample period then use estimated hedge ratios to simulate hedging and estimate hedging effectiveness in a “post-sample” period. This paper derives the statistical properties of the sample-period effectiveness estimator and the statistical properties of the difference between the sample-period and the post-sample period estimators. We find that the bias associated with the sample-period estimator is negligible and that a difference between the sample estimator and the post-sample estimator ties directly to changes in the structural parameters of the hedge-ratio regression. We develop tests for structural change and demonstrate those tests with an empirical example.