摘要
准确厘定费率是科学设计农业保险的重要环节。传统单产风险分析法需要选择特定的概率密度函数或核密度函数,限制了费率厘定的效率和产量风险特征的描述。Markov链预测模型的厘定方法不需要事先设定特定的模型,且统计了正常减产波动范围,因此,能够避免先验性问题。基于此优势,本文利用山东省及其17个地市1993—2020年小麦单产量数据,在验证了Markov链预测模型对小麦单产量预测具有较高的精确度的基础上,进一步运用Markov链预测模型对山东省及其17个地级市的纯费率进行了厘定,结果显示:山东省小麦保险总体费率为1.061%,17个地市的平均费率水平为1.763%。其中,威海市费率最高,为3.897%;青岛市费率最低,为0.980%。研究认为,农业保险的费率厘定应精确拟合单产趋势,灵活结合测算方法,从而实现费率厘定的精确化。
Accurate rate determination is an important link in scientific design of agricultural insurance.The traditional yield risk analysis method needs to select a specific probability density function or kernel density function,which limits the efficiency of rate determination and the description of yield risk characteristics.Since the determination method of the Markov chain prediction model does not need to set a specific model in advance,and the fluctuation range of the normal production reduction is calculated,the priori problem can be avoided.Based on this advantage,this paper uses the wheat yield data of Shandong province and its 17 prefecture-level cities from 1993 to 2020 to verify the high accuracy of the Markov chain prediction model in forecasting the wheat yield.Based on it,the paper further uses the Markov chain prediction model to determine the pure premium rate of Shandong province and its 17 prefecture-level cities.The results show that the overall premium rate of wheat insurance in Shandong province is 1.061%,and the average premium rate level of 17 prefecture-level cities is 1.763%,among which,Weihai has the highest rate of 3.897%,while Qingdao has the lowest rate of 0.980%.The study also believes that the rate determination of agricultural insurance should accurately fit the trend of the unit yield and flexibly combine the calculation methods to achieve the accuracy of rate determination.
作者
杨丰滔
孙乐
陈盛伟
Yang Fengtao;Sun Le;Chen Shengwei
出处
《保险职业学院学报》
2023年第2期30-40,共11页
Journal of Insurance Professional College
基金
山东省金融学会2022年度重点研究课题“碳中和、气候变化与保险业发展”(2022SDJR27)。
关键词
费率厘定
农业保险
生产风险评估
MARKOV链
rate determination
agricultural insurance
production risk assessment
Markov chain
