区域农业产量风险低估的评价研究

A Study on the Underestimated Evaluation of Regional Agricultural Yield Risk

作为我国大力推广的创新型保险产品,区域农业收入保险是对区域农业产量风险与价格风险的双重保障。然而,单产数据空间聚合造成产量风险的低估却是当前农业收入风险评估的一个重要问题,尤其是随着空间区域的增大、产量风险的低估程度就越严重。本文以湖北省黄冈市、咸宁市和十堰市三个地市级水稻产量风险评估为案例,通过两种方法定量评价地市级水稻产量风险的低估程度。方法一是当前最常见的区域产量风险评估方法,直接利用地市级产量数据构建风险评估模型计算产量风险;方法二是收集地市级对应区县产量数据,通过Copula函数构建县级产量损失序列的联合概率分布后,通过Monte Carlo模拟产生区县产量损失序列样本后计算地市级产量风险。通过结果比较发现,方法一获得的三个地级市水稻产量风险较方法二平均低估了51.27%,证明了空间尺度建模不当会带来产量风险评估的严重低估。

As an innovative agricultural insurance product proactively promoted in China,regional agricultural income insurance is a double guarantee for regional agricultural yield risk and price risk.However,the underestimation of agricultural yield risk due to inappropriate spatial scale modeling of unit produce data is an important issue in current agricultural income risk assessment,especially with the increase of spatial areas,the degree of underestimation of yield risk is more serious.Therefore,in this article,taking Huanggang City,Xianning City and Shiyan City of Hubei Province as an example,the underestimation of the city-level rice yield risk was evaluated by two methods.The first method is the most common yield risk assessment method,which directly uses the city-level rice yield data to construct a risk assessment model and calculate the city-level yield risk.The second method is to collect county-level rice yield data of all counties belonging to the city,construct the joint probability distribution of the county-level yield losses through Copula function,and finally calculate the city-level yield risk through Monte Carlo simulation.The results show that the rice yield risk of three cities obtained by the first method is 51.27%lower than that by the second method on average,which fully proves that improper spatial modeling will lead to serious underestimation of yield risk assessment.