Feeder Cattle Futures - The Life of a Contract: Comparing Expiration to Weekly Futures Prices
The Feeder Cattle 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 such as this 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 Feeder Cattle 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 January 1991-May 2009. These weekly averages were then compared to the futures closing price on each contract’s last day of trade.
The weekly prices’ forecast errors were defined as the futures price at expiration minus the futures price in trade week 1, 2, 3, etc, and are expressed as a percentage of the expiration price. A positive error means the weekly price was below the expiration price, indicating under prediction, and a negative error means the weekly price was above the expiration price, indicating over prediction. 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 and variability, as the contract’s end approaches.
Figures 1-8 show the forecast errors of the January, March, April, May, August, September, October, and November contracts over their entire trading periods from 1991-2009. As can be seen, contract errors vary widely. The errors from the August, September, and October contracts tend to be positive, indicating under prediction. The January, March, and April contracts tend to have negative errors, showing over prediction. Finally, errors from the May and November contracts’ vary widely and are both positive and negative throughout. As shown by these figures, each contract’s errors tend to decrease and move toward a 0% error on average as the contracts mature. This is expected because as more information becomes available to traders, they are better able to make pricing decisions.
It is important to know more than the average about the forecast errors. Tables 1-8 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 March, April, and May contracts tended to have more years where the weekly prices were above the expiration price on average, indicating negative errors, or over prediction by the weekly prices. August-November contracts tended to have more years where the weekly prices were below the expiration price, which indicates under prediction. January and December were split between under and over prediction. However, it is important to remember that there are only 19 numbers in each of these averages at most and a large error in any one year can change the averages and standard deviations dramatically.
Table 9 (below) provides analysis on the overall effectiveness of the contracts as predictors of expiration prices. As shown by the very small average errors (.8% of $90/cwt is $.72/cwt), the contracts are very accurate. The most accurate month on average was November. However, the least variable months were January and March.
This analysis is intended to provide some insight into how accurately Feeder Cattle 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.