I know that some of you do not believe in price forecasting. But many of us are engaged in aiding producers to adjust production to demand and in deciding when and how to market; or in aiding distributors and consumers to decide when to buy. We are trying to make some practical use of economic theory and statistics. Whether we acknowledge it or not, we are forecasting or aiding our employers to forecast prices. Enough is known about prices to enable competent analysts to make price forecasts sufficiently accurate to be useful in planning production, marketing, consumption, or buying, to the advantage of the producer, the distributor, or the consumer. Whenever it is possible by such means to bring about a more accurate adjustment of prices to fundamental supply and demand conditions, the producer, the distributor and the consumer may all profit through the elimination of economic wastes. These two propositions are re-phrased from the conclusion of the paper read last year at the meeting of the Farm Economic Association on Progress in Price Analysis and an Appraisal of Success in Price Forecasting. ' I am restating these two propositions as my justification for advocating great care and large expenditures in the collection and analysis of data as to production, prices, and consumption of commodities, for use in price forecasting. The subject assigned to me seems to be comprehensive as to the commodities to be considered, but lack of knowledge of non-agricultural commodities makes it necessary for me to confine my discussion to agricultural commodities.