Forecasting Price Movements: An Application Of Discriminant Analysis

The purpose of this paper is to demonstrate a technique, discriminant analysis, which may be useful in predicting the direction of movement between fall feeder calf prices and spring yearling prices. The results of the discriminant analysis model are then compared with a conventional regression approach in terms of relative accuracy of predictions. The usefulness of incorporating the direction of price movement as a variable in a price prediction model is also evaluated. Generally, the results suggest that the discriminant analysis approach provides useful information, and the directional variable improves forecasts when incorporated into a traditional forecasting model.