Cyclical fluctuations in prices and production have long characterized the United States hog industry. Recent evidence suggests that the length of the hog cycle has changed. In order to determine whether the change in cycle length is statistically significant, the bootstrap technique is employed to derive confidence intervals for point estimates of the hog cycle. Application of the bootstrap technique to time series models is discussed and empirical results are presented. It is concluded that the hog cycle is undergoing rather complicated changes based on cycle lengths that are calculated to be statistically different from zero.
