An Analysis of Ill‐Posed Production Problems Using Maximum Entropy
展示如何用最大熵方法从极有限的数据中恢复灵活的成本函数，并论证数学规划与传统计量技术之间存在基于可用信息的连续分析谱系。
Abstract Production economics problems are often ill‐posed. This means that the number of parameters to be estimated is greater than the number of observations. In this article we show how to recover flexible cost functions from very limited data sets using a maximum entropy approach. We also argue that there exists a continuum of analysis between mathematical programming and traditional econometric techniques which is based solely upon the available information. The limiting case of a multi‐output cost function recovered using only a single observation of a farmer's allocation decisions can be easily extended to handle more than one observation.