A reappraisal of the forecasting performance of corn and soybean new crop futures

Abstract

ESO 2310This analysis evaluates the forecasting ability of the December corn futures contract and November soybean futures contracts during the previous spring. A regression equation is estimated which accounts for the well-known non-stationarity of commodity prices over the period 1952-1995. Results of this regression imply that the spring-time quotes of the corn and soybean harvest futures contracts are unbiased estimates of the prices at harvest. In addition, since 1974 the spring-time quotes are able to significantly predict the harvesttime quotes. This finding implies that farmers and others can use harvest futures at planting as a source of information concerning prices at harvest. Furthermore, in accordance with Stein, because the futures contracts are unbiased forecasts of realized prices, then the corn and soybean futures markets are functioning well in the sense that only unavoidable social loss exits. Magnitude of the unavoidable loss is measured by R 2 • R 2 increased for corn but decreased for soybeans between 1952-72 and 1974-95. These findings suggest that unavoidable social loss has decreased in the corn market, but has increased in the soybean market. Last, since 1973, the spring quotes of the corn and soybean harvest futures were significantly less variable from year-to-year than the eventual harvest prices. This finding suggests that hedging expected corn and soybean production at planting can significantly reduce year-to-year variability in price received at harvest.

Market efficiency in agricultural futures markets

“…For example, Fama and French (1987) and, more recently, Zulauf et al (1999) speci®ed Equation 3 alternatively as…”

A combination method for interval forecasting of agricultural commodity futures prices

“…But there exists a much smaller body of research examining the predictability of agricultural commodity futures prices [17][18][19]. Furthermore, an important point to note from past studies [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] mentioned above is their preoccupation with point forecasting rather than interval one.…”

“…First, as was mentioned above, although extensive mounts of methods [1][2][3][4][5][6][7][8][9][10][11][12][14][15][16][17][18][19] have been developed for futures prices modeling and forecasting, most of them rely only on single-valued futures price series. The interval forecasting of futures prices, particularly agricultural commodity futures prices, has not been widely explored.…”

“…Second, by introducing VECM and MSVR, the well-established ''linear and nonlinear'' time series modeling framework can be extended to deal with ITS forecasting, which can fully take the advantages of the unique strength of VECM and MSVR in linear and nonlinear modeling, respectively. Third, the most published literature on agricultural commodity futures prices forecasting focuses on international markets, and on the US market in particular [17][18][19]. There is limited research on the predictability of Chinese agricultural commodity futures market.…”

Market efficiency in Indian commodity futures markets

“…Hansen and Hodrick (1980), Hakkio (1981) and Baillie et al (1983) used difference equation to test the market efficiency and unbiasedness hypothesis. Fama and French (1987) and Zulauf et al (1999) regressed spot price changes (Δ S t ) on the lagged basis ( F t −1 − S t −1 ) to test the same.…”