基于GARCH-Copula模型的国际原油价格与可再生能源股价相关性研究
发布时间：2018-10-23 15:00
【摘要】：进入二十一世纪以来,随着化石能源的日益枯竭以及人们对空气环境质量要求的日益提高,可再生能源走进了人们的视野,并且在过去的十几年里得到了快速的发展。根据国际能源署披露的数据表明,2014年全球可再生能源领域投资金额达2.7万亿美元。而可再生能源作为例如石油、煤炭等传统能源的替代品,其发展或多或少的受到国际原油价格的影响。因此,就很有必要研究国际原油价格与可再生能源股价之间的相关关系,以帮助投资者更好的进行可再生能源的投资决策。本文通过使用Copula函数描述国际原油价格与可再生能源股价之间的相关结构,建立了 GARCH-Copula模型。并且回顾了过往对国际原油价格、可再生能源及其二者相关性的研究,介绍了传统相关性度量方法及其存在的一些缺点,提出了使用Copula函数能够更好的度量金融时间序列之间相关性的观点,比较详细的介绍了 Copula函数的定义、性质、分类以及与相关性度量方法的关系,最后提出引入Copula函数运用蒙特卡洛模拟的方法计算投资组合VaR。本文选取2009年12月31日至2016年12月31日布伦特原油价格、Wind新能源指数、标普全球清洁能源指数以及WilderHill清洁能源指数共1648组数据作为研究对象进行实证研究。文章首先对四个收益率序列的边缘分布进行t-GARCH(1,1)建模,之后选取了正态 Copula 函数、Gumbel Copula 函数、Clayton Copula函数以及Frank Copula函数对布伦特原油与三个可再生能源指数之间的相关结构进行建模,最后根据最优Copula函数计算了尾部相关系数及投资组合的VaR,并且确定最优投资组合。
[Abstract]:Since 21 century, with the increasing depletion of fossil energy and the increasing demand for air quality, renewable energy has come into people's vision, and has been rapidly developed in the past ten years. Investment in renewable energy globally reached $2.7 trillion in 2014, according to data released by the International Energy Agency. As a substitute for traditional energy such as oil and coal, the development of renewable energy is more or less affected by the international crude oil price. Therefore, it is necessary to study the correlation between the international crude oil price and the share price of renewable energy, so as to help investors make better investment decisions on renewable energy. In this paper, the GARCH-Copula model is established by using Copula function to describe the correlation structure between the international crude oil price and the share price of renewable energy. And reviewed the past research on the international crude oil price, renewable energy and their correlation, introduced the traditional correlation measurement methods and some shortcomings, This paper puts forward the viewpoint that using Copula function can better measure the correlation between financial time series, and introduces in detail the definition, property, classification of Copula function and the relationship between Copula function and correlation measurement method. Finally, the paper introduces Copula function to calculate portfolio VaR. by Monte Carlo simulation. From December 31, 2009 to December 31, 2016, Brent crude oil price, Wind new energy index, S & P global clean energy index and WilderHill clean energy index were selected for empirical research. In this paper, the edge distribution of four return sequences is first modeled by t-GARCH (1 / 1). Then the normal Copula function, Gumbel Copula function, Clayton Copula function and Frank Copula function are selected to model the correlation structure between Brent crude oil and three renewable energy indices. Finally, according to the optimal Copula function, the tail correlation coefficient and the VaR, of the portfolio are calculated and the optimal portfolio is determined.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F416.22;F831.51;F764.1
，
本文编号：2289591
[Abstract]:Since 21 century, with the increasing depletion of fossil energy and the increasing demand for air quality, renewable energy has come into people's vision, and has been rapidly developed in the past ten years. Investment in renewable energy globally reached $2.7 trillion in 2014, according to data released by the International Energy Agency. As a substitute for traditional energy such as oil and coal, the development of renewable energy is more or less affected by the international crude oil price. Therefore, it is necessary to study the correlation between the international crude oil price and the share price of renewable energy, so as to help investors make better investment decisions on renewable energy. In this paper, the GARCH-Copula model is established by using Copula function to describe the correlation structure between the international crude oil price and the share price of renewable energy. And reviewed the past research on the international crude oil price, renewable energy and their correlation, introduced the traditional correlation measurement methods and some shortcomings, This paper puts forward the viewpoint that using Copula function can better measure the correlation between financial time series, and introduces in detail the definition, property, classification of Copula function and the relationship between Copula function and correlation measurement method. Finally, the paper introduces Copula function to calculate portfolio VaR. by Monte Carlo simulation. From December 31, 2009 to December 31, 2016, Brent crude oil price, Wind new energy index, S & P global clean energy index and WilderHill clean energy index were selected for empirical research. In this paper, the edge distribution of four return sequences is first modeled by t-GARCH (1 / 1). Then the normal Copula function, Gumbel Copula function, Clayton Copula function and Frank Copula function are selected to model the correlation structure between Brent crude oil and three renewable energy indices. Finally, according to the optimal Copula function, the tail correlation coefficient and the VaR, of the portfolio are calculated and the optimal portfolio is determined.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F416.22;F831.51;F764.1
，
本文编号：2289591
本文链接：https://www.wllwen.com/kejilunwen/yysx/2289591.html
