Abstract The study sought to explain the economic structure of the short-run market for redwood lumber and to prepare short-range forecasts of price, new orders, shipments, production, stocks, and unfilled orders. Major features of the redwood lumber industry were studied and short-run market theory was developed. The market structure was described by a system of linear equations explaining each of the six characteristics studied. The recursive nature of the system made it acceptable from a statistical point of view to use direct least squares to estimate regression coefficients. Elasticity estimates were used to compare the influence of individual explanatory variables. Forecasts were generated using estimated structural, solved structural, and reduced form equations. Greatest success was achieved in forecasting price and stocks. Results were poorest with regard to new orders. No forecasting method was clearly better than the others.
William McKillop (Sun,) studied this question.