基于风险分区的福建省森林火灾与病虫害联合损失分布的研究
|
摘要
以福建省98个市(县、区)2000～2015年的森林火灾与病虫害损失情况为研究对象,通过对福建省进行森林火灾与病虫害风险灾害分区,把福建省分为高、中、低三个风险区。在此基础上,对不同的风险区根据其森林火灾与病虫害损失情况,分别拟合相应单变量边缘分布函数。最后利用合适的Copula模型将每个风险区森林火灾与病虫害边缘分布与其相关系数连接,从而构造出各风险区二维联合损失分布模型。结果表明:福建省高风险区森林火灾与病虫害可用Gumbel-Copula函数来拟合,其上尾相关性为0.548,相关性较高,协同发生的概率较大;而中风险区可使用Frank Copula函数拟合较理想,其上尾相关性为0.387,下尾相关性为0.411,呈对称分布。森林火灾与病虫害联合损失分布的研究不仅为保险公司制定有效的森林综合保费提供了理论依据,同时也为福建省有效加强森林火灾与病虫害治理提供了决策参考。
Taking the situation of forest fire and loss of forest pests and diseases in 98 cities(counties and districts) of Fujian Province from 2000 to 2015 as the research object, this paper dividesd Fujian Province into three risk zones:high, medium and low by dividing Fujian Province into forest fire and pest risk zones. On this basis, according to the occurrence of forest fire and pest losses in different risk areas, the corresponding single variable edge distribution functions were fitted. Finally, the appropriate Copula model was used to connect the forest fire and the edge distribution of pests and diseases in each risk area with their correlation coefficients, so as to construct a two-dimensional joint losses distribution model for each risk area. The results showed that: the Gumbel-Copula function can be used to fit forest fires and pests in high-risk areas of Fujian Province. The upper-tail correlation is 0.548, the correlation is high, and the probability of synergistic occurrence is high. The Frank Copula function can be used to fit the middle-risk areas. The upper-tail correlation is 0.387, and the lower-tail correlation is 0.411 showing a symmetrical distribution. The study on joint losses distribution of forest fire and pests not only provides a theoretical basis for insurance companies to formulate effective comprehensive forest insurance premiums, but also provides decision-making reference for Fujian Province to effectively strengthen forest fire and pests management.
Taking the situation of forest fire and loss of forest pests and diseases in 98 cities(counties and districts) of Fujian Province from 2000 to 2015 as the research object, this paper dividesd Fujian Province into three risk zones:high, medium and low by dividing Fujian Province into forest fire and pest risk zones. On this basis, according to the occurrence of forest fire and pest losses in different risk areas, the corresponding single variable edge distribution functions were fitted. Finally, the appropriate Copula model was used to connect the forest fire and the edge distribution of pests and diseases in each risk area with their correlation coefficients, so as to construct a two-dimensional joint losses distribution model for each risk area. The results showed that: the Gumbel-Copula function can be used to fit forest fires and pests in high-risk areas of Fujian Province. The upper-tail correlation is 0.548, the correlation is high, and the probability of synergistic occurrence is high. The Frank Copula function can be used to fit the middle-risk areas. The upper-tail correlation is 0.387, and the lower-tail correlation is 0.411 showing a symmetrical distribution. The study on joint losses distribution of forest fire and pests not only provides a theoretical basis for insurance companies to formulate effective comprehensive forest insurance premiums, but also provides decision-making reference for Fujian Province to effectively strengthen forest fire and pests management.
引文
崔巍.林火与病虫害的相互作用[J],世界林业研究,2010,23(5):34~38
贺昌彩.浅析森林病害的综合防治措施[J].绿色科技,2016(1):62~63
黄生志,黄强,王义民,等.基于SPI的渭河流域干旱特征演变研究[J].自然灾害学报,2015,24(1):15~22
刘东.南方主要集体林区森林火灾风险分区模型研究[D],北京,北京林业大学,2014
史道济,姚庆祝.改进Copula对数据拟合的方法[J].系统工程理论与实践,2006(4):49~55
翁小艳.条件阿基米德Copula[D].成都:西南交通大学,2008
吴丽红.对于森林虫害的综合防治分析[J].新农村黑龙江,2017(9):166
张世英,韦艳华,张世英.金融市场的相关性分析--CopulaGARCH模型及其应用[J].系统工程,2004,22(4):7~12
张尧庭.连接函数(copula)技术与金融风险分析[J].统计研究,2002(4):48~51
Dean Fantazzini.Copulas conditional dependence measures for portfolio management and value atrisk[J].Working paper,2003,66(15):88~103
Forbes K,Rigobon R.No contagion,only independence:measuring stock market co-movement[J],Journal of Finance,2002,57(5):223~261
Nelson R B.An introduction to Copulas:lecture notes in statistics[M].New York:Springer,1998
Sklar A.Fonctions de repartition an dimensions et leurs marges[J].Publication de Institut de Statistique de Universite de Paris,1959(8):229~231
贺昌彩.浅析森林病害的综合防治措施[J].绿色科技,2016(1):62~63
黄生志,黄强,王义民,等.基于SPI的渭河流域干旱特征演变研究[J].自然灾害学报,2015,24(1):15~22
刘东.南方主要集体林区森林火灾风险分区模型研究[D],北京,北京林业大学,2014
史道济,姚庆祝.改进Copula对数据拟合的方法[J].系统工程理论与实践,2006(4):49~55
翁小艳.条件阿基米德Copula[D].成都:西南交通大学,2008
吴丽红.对于森林虫害的综合防治分析[J].新农村黑龙江,2017(9):166
张世英,韦艳华,张世英.金融市场的相关性分析--CopulaGARCH模型及其应用[J].系统工程,2004,22(4):7~12
张尧庭.连接函数(copula)技术与金融风险分析[J].统计研究,2002(4):48~51
Dean Fantazzini.Copulas conditional dependence measures for portfolio management and value atrisk[J].Working paper,2003,66(15):88~103
Forbes K,Rigobon R.No contagion,only independence:measuring stock market co-movement[J],Journal of Finance,2002,57(5):223~261
Nelson R B.An introduction to Copulas:lecture notes in statistics[M].New York:Springer,1998
Sklar A.Fonctions de repartition an dimensions et leurs marges[J].Publication de Institut de Statistique de Universite de Paris,1959(8):229~231