基于g-h分布的我国地震保险损失拟合研究

发布时间：2018-02-14 10:05

本文关键词： g-h分布 g-hVaR 地震保险　出处：《中国海洋大学》2013年硕士论文　论文类型：学位论文

【摘要】：我国是世界上地震灾害最为严重的地区之一。地震自然灾害的发生,往往会给人们造成巨大的财产损失,严重影响我国经济的稳定、健康、快速发展。虽然地震频发,但是我国的地震保险机制却是相当不完善,我国地震灾害的经济损失和地震保险赔付之间存在着巨大反差。面对巨大的地震损失,单纯依靠政府补贴和社会捐助很难满足灾后重建所需的巨额资金。地震保险是经济社会中最为重要的风险分散和经济补偿手段,对经济社会的稳定发展具有重要作用。积极推动地震保险制度建立和完善,是减轻政府财政压力,提高抗震救灾能力的有效手段。地震保险制度的完善,除了从政策层面上来加以引导解决外,还要利用地震保险精算对其提供技术支持。地震保险损失分布是地震保险精算的基础和核心,也是保险产品定价的基础,是合理提取风险准备金的前提,是保险监管的重要依据。所以,合理的拟合地震保险损失分布具有重要的理论和现实意义。基于这一前提,本文引入了g-h分布。g-h分布目前大多被用于金融领域,用来研究投资收益率的特征,很少被应用于保险领域。本文详细介绍了g-h分布并将其应用到保险领域对地震灾害的损失分布进行拟合。主要内容分为以下四个部分：首先,本文介绍了地震保险以及开展地震保险的意义,并对国内外地震保险的发展状况进行了系统的梳理,为下文研究奠定基础。其次,发展地震保险的一个重要因素是其精算研究,而地震保险精算的前提是能够较好的拟合地震损失分布。本文介绍了常用的分布模型,并分析了他们对于拟合具有尖峰厚尾特征的地震损失分布中存在的不足,由此引入了g-h分布。再次,本文详细介绍了g-h分布。g-h分布是在正态分布的基础上,增加了能够描述分布的偏斜特性g和尾部特性的参数h,因此能够很好地刻画地震巨灾损失数据尖峰厚尾的特征。本文依次介绍了g-h分布的定义、性质、参数的分位数估计方法,以及基于g-h分布的VaR的求法。在此基础上,本文探讨了g-h分布在保险领域的应用前景。最后,论文基于以上研究,选取了1969年至2011年期间超过1亿的地震灾害的损失数据进行实证分析。结果表明,g-h分布对于地震损失数据具有良好的拟合效果。本文的研究重点及创新之处在于：引入了g-h分布来分析地震风险,并实证了其适用性；拟合了我国地震风险的损失分布,为地震保险的精算定价做了准备工作；介绍了基于g-h分布的VaR方法并在实证中运用；在前面的研究基础上对g-h分布在地震保险中的运用前景进行了探讨。
[Abstract]:China is one of the areas with the most serious earthquake disasters in the world. The occurrence of earthquake natural disasters often causes huge property losses to people and seriously affects the stability, health and rapid development of our economy. Although earthquakes occur frequently, However, the mechanism of earthquake insurance in our country is not perfect. There is a huge contrast between the economic loss of earthquake disaster and the indemnity of earthquake insurance in China. Relying solely on government subsidies and social donations is hardly enough to meet the huge amount of money needed for post-disaster reconstruction. Earthquake insurance is the most important means of risk dispersion and economic compensation in an economic society. It plays an important role in the steady development of economy and society. Actively promoting the establishment and improvement of earthquake insurance system is an effective means to lighten the financial pressure of the government and improve the ability of earthquake relief and relief. The perfection of the earthquake insurance system should not only be guided from the policy level, but also provide technical support by means of the earthquake insurance actuarial. The distribution of earthquake insurance losses is the foundation and core of the earthquake insurance actuarial system. It is also the basis of the pricing of insurance products, the premise of reasonable drawing risk reserve and the important basis of insurance supervision. Therefore, it is of important theoretical and practical significance to fit the distribution of earthquake insurance losses reasonably. In this paper, the g-h distribution. G-h distribution is mostly used in the field of finance, which is used to study the characteristics of the return on investment. The g-h distribution is introduced in detail and applied to the field of insurance to fit the loss distribution of earthquake disaster. The main contents are as follows:. Firstly, this paper introduces the significance of earthquake insurance and the development of earthquake insurance, and systematically combs the development of earthquake insurance at home and abroad, which lays a foundation for the following research. Secondly, an important factor in the development of earthquake insurance is its actuarial research, and the premise of earthquake insurance actuarial is to fit the distribution of earthquake loss better. Their shortcomings in fitting the seismic loss distribution with the characteristics of peak and thick tail are analyzed, and the g-h distribution is introduced. Thirdly, this paper introduces in detail that g-h distribution. G-h distribution is based on normal distribution. In this paper, the definition, property and quantile estimation method of g-h distribution are introduced, which can describe the skew characteristic g of distribution and the parameter h of tail characteristic, so it can describe the feature of peak and thick tail of seismic catastrophe data. In this paper, the definition, property and quantile estimation method of g-h distribution are introduced in turn. On the basis of this, the application prospect of g-h distribution in the field of insurance is discussed. Finally, based on the above research, we select the loss data of more than 100 million earthquake disasters from 1969 to 2011 for empirical analysis. The results show that the g-h distribution has a good fitting effect on the seismic loss data. The key points and innovations of this paper are as follows: introducing g-h distribution to analyze earthquake risk and demonstrating its applicability, fitting the loss distribution of earthquake risk in China, and preparing for actuarial pricing of earthquake insurance. This paper introduces the VaR method based on g-h distribution and its application in practice, and discusses the application prospect of g-h distribution in earthquake insurance based on the previous research.
【学位授予单位】：中国海洋大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F842.64

【相似文献】