Can a rational expectation storage model explain the USDA ending grain stocks forecast errors?
研究了理性预期库存模型在多大程度上能解释美国农业部对玉米、大豆和小麦的期末库存预测误差，发现预测误差部分可由时间基差和月度库存解释，改进幅度分别为7.1%、7.2%和4.5%。
Abstract We examine whether, and to what extent, a rational expectation storage model can be used to explain the US Department of Agriculture (USDA) ending stocks forecast errors for corn, soybean, and wheat. We show there is a predictable component of forecast errors from 1980/81 to 2018/19 that can partially be explained by the temporal basis and monthly stocks, two important factors from a rational expectation storage model that codetermines carry. The USDA ending stocks forecasts can be improved by 7.1%, 7.2%, and 4.5% for corn, soybean, and wheat, respectively. The improvement is most significant when the last year's carry is small.