Techniques for Forecasting Farm and Industrial Tractor Sales

Various techniques are presented and evaluated for their ability to forecast equipment sales in the agricultural industry. The forecasting methods discussed include trend approximation, time series moving averages, exponential smoothing, Delphi Method, buyer intentions and econometric models. A model that predicts future sales for industrial tractors was presented including a 1978 forecast. Leading indicators for Industrial Tractor Sales were GNP construction dollars, index of prices of construction machinery and new housing starts. Also, a farm tractor sales model and the resulting difficulty in finding leading indicators were discussed. Statistical methods required for econometric forecasting and model validity testing are given.