Hog Futures - The Life of a Contract Comparing Expiration to Weekly Futures Prices
The Live and, more recently, Lean Hog futures market is a single location where anyone with an opinion on what prices will be in the future can essentially vote their forecast. The resulting futures prices represent a “composite” forecast at a particular point in time. However, futures markets trade on known information and react as new information becomes available. Research has repeatedly shown that the futures are as accurate, or better, than other forecasting methods, but just how good of a predictor of the contract expiration price are weekly futures prices?
This simple analysis compares the Live/Lean weekly futures prices to their contract’s expiration price in order to evaluate their accuracy. Weekly prices were an average of the futures closing prices, Monday through Friday, for each week of the contract, from 1990-2008. These weekly averages were then compared to the futures closing price on each contract’s last day of trade.
The weekly prices’ forecast error was defined as the futures price at expiration minus the futures price in trade week 1, 2, 3, etc. A positive error means the weekly price was below the expiration price, and a negative error means the weekly price was above the expiration price. Because more information becomes available to futures traders as the contract matures, we would expect the weekly prices to inch closer to the expiration price, decreasing their error, as the contract’s end approaches.
The weekly price’s forecast errors are measured as a percent of the futures price at expiration. Figures 1-7 show the forecast errors of the February, April, June, July, August, October, and December contracts over their entire trading periods from 1990-2008. As can be seen, contract errors vary widely. February and October vary both above and below 0% error. April, June, and July tend to follow the same pattern of starting with positive errors that fall into negative errors, which then increase again and settle back around 0% as the contract matures. August, on average, has positive errors throughout the contract, while December has negative errors throughout. These figures also show that the range which each contract’s errors take decreases considerably as the contracts mature. All contracts end closer to a 0% error on average than when they started. This is expected as more information becomes available to traders.
It is important to know more than the average about the forecast errors. Tables 1-7 below report the average and standard deviation of the errors, and number and percentage of years each weekly price was above or below the expiration price. Standard deviation is a measure of variability around the average, and under normal conditions the actual forecast is expected to be within plus or minus one standard deviation of the average approximately two-thirds of the time. A larger standard deviation indicates more variation in the error. With all months, the variation in the errors tends to become dramatically smaller as expiration approaches, indicating that the accuracy of the weekly prices increases closer to expiration. Years above and below again show the variation each weekly price takes from the expiration price. The contracts for April through October tended to have more years where the weekly prices were below the expiration price on average, indicating positive errors, or under prediction by the weekly prices. February varied widely, and December had more years with weekly prices above the expiration price on average, indicating over prediction. However, it is important to remember that there are only 19 numbers in each of these averages and a large error in any one year can change the average dramatically. Such could be the case with December and its 1998 contract which was extremely negative for the entire period.
Table 8 (below) provides analysis on the overall effectiveness of the contracts as predictors of expiration prices. As shown by the very small average errors, and the overall average error of 0.0%, the contracts are very accurate. The most accurate month on average was February, which also has one of the lowest standard deviations. The least variable month was August. December’s forecast error and variability are both significantly larger than other months on average. Again, it’s likely this was caused by its 1998 contract.
This analysis is intended to provide some insight into how accurately Lean Hog futures predict the contract expiration price. The results of this simple analysis suggest that they are very accurate, and that as increasing amounts of information become available, weekly futures prices become increasingly accurate at predicting expiration prices. This is shown by the errors’ tendencies to approach zero and the decreases in their standard deviations as the contract matures. As is the case in all economic situations, more information is always beneficial, and helps traders make more accurate and profitable decisions.