林业经济问题2025，Vol.45Issue(1)：101-112,12.DOI:10.16832/j.cnki.1005-9709.20240559
基于Copula模型的江西省森林综合保险纯费率厘定的研究
The Pure Premium Rate Determination of Forest Comprehensive Insurance in Jiangxi Province Based on Copula Model
摘要
Abstract
(1) Background——In order to disperse the risks of various forest disaster losses,the forest insurance sys-tem has gradually developed into a forest comprehensive insurance system covering multiple liabilities such as fires,forest pests disasters and natural disasters.Due to the correlation between forest fires and pests losses varia-bles,the probability of occurrence and its loss distribution of forest comprehensive risks should be considered when determining the pure premium rate of forest comprehensive insurance.However,there is little literature on the application of cross-impact assessment methods in the study of premium rate determination of forest compre-hensive insurance. (2) Methods——The forest risk coefficients were determined by constructing the indexes system of forest comprehensive risk assessment and cluster analysis,and the Copula model was selected to construct the compre-hensive loss rate function of forest disasters,and then the pure premium rate of forest comprehensive insurance of prefecture-level cities in Jiangxi Province was determined.The data on the occurrence and control of forest pests and diseases and forest fires were obtained from"China Forestry and Grassland Statistical Yearbooks"and"Chi-na Statistical Yearbooks"from 2003 to 2022,respectively.The forest resources data were from the national forest resources inventory data,and the data of 11 prefecture-level cities in Jiangxi Province were from"Jiangxi Statisti-cal Yearbooks"from 2009 to 2016. (3) Results——First,the comprehensive risk levels of forest fires and forest pests and diseases in Jiangxi Province can be divided into three risk areas.The high-risk areas are Nanchang City,Jingdezhen City,Pingxiang City and Jiujiang City,the medium-risk areas are Xinyu City,Yingtan City,Ji'an City,Yichun City,Fuzhou City and Shangrao City,and the low-risk area is Ganzhou City,whose risk coefficients are 1.4,1.2 and 1.0,re-spectively.Second,univariate forest fires loss distribution conforms to gamma distribution,univariate forest pests and diseases loss distribution conforms to log-normal distribution.There is a negative correlation between forest fires loss and forest pests and diseases loss,and the tail correlation is weak.Frank-Copula function shows a higher descriptive ability in describing the structure of forest disasters loss samples.Third,under the condition of deductible coefficient and compensation ratio coefficient of 1.0,the pure premium rate results of forest compre-hensive insurance of 11 prefecture-level cities are significantly different. (4) Conclusions and Discussions——Based on the above research conclusions,the following policy sug-gestions are put forward:First,the economically developed areas in the north of Jiangxi Province should increase forestry investment and pay attention to forest disaster prevention.Second,implement forest insurance premium rate determination at a smaller regional scale.Division of risk areas and differential premium rate determination of forest insurance in a smaller scale are conducive to the development of forest insurance in Jiangxi Province,im-prove the enthusiasm of insurance companies to underwrite and renew products,and also help to motivate forestry farmers,so that insurance holders pay attention to the prevention of forest disasters.Third,continuously optimize the comprehensive forest insurance.Expand the scope of responsibility of forest comprehensive insurance,explore a more scientific and refined premium rate determination method,determine the appropriate insurance amount,insurance premium rate and subsidy ratio.关键词
Copula模型/聚类分析/森林综合灾害损失率/风险系数Key words
copula model/cluster analysis/forest comprehensive disaster loss rate/risk coefficients分类
管理科学引用本文复制引用
王淼,石焱,胡明形,邵杨,周文琪,卢妍洁..基于Copula模型的江西省森林综合保险纯费率厘定的研究[J].林业经济问题,2025,45(1):101-112,12.基金项目
国家林业和草原局委托项目(JYC-2024-0012) （JYC-2024-0012）
