Forecasting price movements: An application of discriminant analysis

Abstract

The purpose of this paper is to demonstrate a technique, discriminant analysis, which might be useful in predicting the direction of movement of fall feeder calf prices to spring yearling prices. The discriminant analysis approach is then compared to regression in predicting the direction of price movement. Finally, the usefulness of incorporating the direction of price movement as a variable in a price prediction model is evaluated. Results suggest that the discriminant analysis approach and the prediction and use of the direction of price movement as a variable in forecasting models provides useful information and improves forecasts.