Information entropy method of crop yield distributions: Implications for pricing crop insurance contracts

Crop yield distributions are the basis of premium rating of agricultural insurance. In this paper, we use maximum entropy density estimation procedures to evaluate crop yield distributions. The procedures are applied to a data set of crop yields for rice, corn, soybean and peanut, for the period of 1980 through 2010 Liaoning province in China. Meanwhile, based on the maximum entropy distributions, implications for rating crop insurance contracts are discussed. Further, all other alternative premium rates are calculated and compared with the maximum entropy ones. The results show that the maximum entropy optimization model can consider more information about crop yield distributions. So, in the situation of limited information, the maximum entropy premium rates are alternative. This study provides a more reasonable method of the determination of agricultural insurance rate which is helpful to the scientific decision-making of agricultural risk.