Crop decision planning is an important part of effective farm management. Because of the many uncertain factors such as weather variations, technology advances, and crop yields and prices, all of which prove to change considerably, decision planning can be very complex. The focus of this paper is to develop accurate yield and price forecasting models to aid in decision planning. For yield forecasting, we establish a crop-weather model using regression methods. We propose a semiparametric regression model, which accounts for both within- and between-year relationships in the data. For price forecasting, we develop a futures-based model, which predicts the cash price from futures price and basis. We estimate the yearly basis pattern using a functional model-based approach and adjust the basis estimates with the futures prices to forecast the crop cash price. In both forecasting models, we estimate confidence bands for the predictions that can be further integrated in the decision planning model. Key words and phrases: functional principal component analysis, functional linear regression, p-splines, functional model-based clustering, crop decision planning, yield forecasting, price forecasting. 1

this document for non-commercial purposes by any means, provided that this copyright notice appears on all such copies.

The statistical forecasting efficiency of new crop corn and soybean futures is the topic of frequent academic inquiry. However, few studies address the usefulness of these forecasts to economic agents ’ decision making. Each year Central Illinois producers are faced with the decision to plant either corn or soybeans on marginal acreage. Agronomic concerns aside, these decisions hinge on the expected relative return of corn versus soybeans, which is largely a function of expected new crop prices. Do new crop futures prices reliably guide producers into the correct production decision? The results suggest that over the entire period of the analysis, futures markets provide only marginal decision-making information to the producer; however, more recent signals do appear to be useful. Further analysis explores several possible factors that could explain why the signals have improved so significantly since 1985. The Forecasting Value of New Crop Futures: A Decision-Making Framework

With volatile grain markets, rapidly increasing average size of U.S. cash grain farms, and narrow profit margins due to rising costs for land and other inputs, grain producers have become increasingly interested in expanding their marketing window by utilizing pre-harvest pricing

An efficient futures market should provide a forecast of the future spot price which reflects all publicly available information; ideally, for effective price discovery such forecasts would also be unbiased. The trading of maize futures contracts began in South Africa (SA) in mid-1996 after r-the power of the Maize Board to set maize producer prices was abolished. Cointegration analysis of the efficiency ofSA white and yellow maize futures markets shows that (1) the futures price for white maize was a biased predictor of the spot (cash) price for white maize in 1997, but an unbiased predictor in 1998 (evidence of a market learning process), and (2) the futures price for yellow maize was an unbiased predictor of the spot (cash) price for yellow maize in 1997 and 1998. White maize is predominantly used for human consumption and SA is considered a leader in the world market for white maize. Yellow maize is mostly used for animal consumption and is traded internationally on the Chicago Board of Trade in the United States of America. This makes the domestic futures price for yellow maize more susceptible to international maize marketing conditions than the domestic futures price for white maize. The relatively greater volume of trade in yellow maize would provide more reliable information about crop and market

Three procedures are used to test Fama semistrong form efficiency of harvesttime price of Kansas City July wheat futures from 1947 through 1995. The three methods are (a) testing for jointly significant parameter estimates on nonfutures explanatory variables in econometric forecasting models, (b) testing the relative accuracy between model-based forecasts and using deferred futures prices as forecasts, and (c) testing for abnormal profits associated with simulated futures trading signaled by the fore-casts. Kansas City July wheat futures are generally efficient. Furthermore, relative to the efficiency associated with forecasts constructed one to two months before harvest, the efficiency associated with the five- to six-month period before harvest has in-creased, especially since the early 1980s. Key words: futures efficiency, futures trading, wheat futures