农作物产量保险区域化差别费率厘定的可行性——基于非参数核密度估计实证

采用非参数核函数平滑法以辽宁省、黑龙江省以及大连市的水稻、玉米和大豆三种农作物历年单位面积产量为例拟合了单产损失分布,同时利用传统的正态概率密度对区域作物单产分布进行了拟合。在拟合损失分布的基础上,分别厘定出不同保险水平农作物区域产量保险的纯保险费率。经测算发现,传统的正态概率密度下厘定的纯保险费率均低于非参数核密度下测算的纯费率,正态法低估了农作物单产的风险。保险水平在70%⁸0%间的参数法及非参数法测算的纯保险费率均低于政策性农业保险的现行费率。另外,在数据可得的基础上,还应该确定适当的厘定保费费率的区域以充分识别风险,更精确的计算保费。

The Feasibility of Insurance Regionalization Determined by the Difference in Rates for the Crop Yields:Based on the Estimate Empirical of Non-parametric Kernel Density

In this paper, by using the non-parametric nuclear function smoothing method and based on the calendar year of per unit area for three crop yields ：rice, corn and soybean in Liaoning Province, Heilongjiang Province and Dalian City to detect the distribution of the yield loss, at the same time, fitted the normal probability density distribution of regional crop yields. On the basis of determination of the loss distribution, respectively determining the pure premium rate for ＂the different level of insurance crop area yield. The determination found that the pure premium rate determined under the normal probability density is lower than that the non-parametric kernel density and the normal method underestimated the risk of crop yield . The parameter method of which insurance levels between 70 to 80% and non- parametric method estimates the pure premium rates were lower than the current rate of policy-oriented agricultural insurance. In addition, we also believe that on the basis of availability of data, we should determine the area of appropriate premium rates in order to fully identify the risks andmake a more accurate calculation of premiums.