摘要：Most of the papers published in Agricultural Economics Research originate either within the Bureau of Agricultural Economics itself or in projects carried on cooperatively with the State Agricultural Colleges. This is not an inviolable rule, however, and this issue contains two exceptions to it: the present article and another, farther on, by Messrs. Todd and Zirkle of the Forest Service. In both cases, of course, the subject matter appears to be of especial interest to workers within the Bureau. This article summarizes a study which was planned and carried forward by Messrs. Baker and Paarlberg as a part of Baker's graduate work at Purdue. Several of us in the Bureau were acquainted with the fact that the study was under way and we not only supplied Mr. Baker with some of his Agricultural Situation references which were used as the source of the forecasts but also had some opportunity to discuss concepts and methods. The subject matter is of such a direct and immediate interest to all those interested in agricultural outlook work that we believed it warranted an invitation too Messrs. Baker and Paarlberg to prepare this article. As was indicated in the forward to James P. Cavin's "Forecasting the Demand for Agricultural Products" in the July issue, we need more appraisal and criticism of our forecasting results rather than less; and this is in effect a companion to Cavin's analysis, as this article evaluates the accuracy of our forecasting results while Mr. Cavin's article dealt chiefly with methods by which the forecasts are derived.
