国内外原油市场的多重分形性与交互相关性分析

发布时间：2018-01-08 07:07

  本文关键词：国内外原油市场的多重分形性与交互相关性分析　出处：《南京财经大学》2014年硕士论文　论文类型：学位论文

  更多相关文章： 原油市场 收益率序列 多重分形性 交互相关性 多重分形谱

【摘要】：随着我国对国际原油进口量的增大,国内的原油价格对国际的原油价格存在很大的依附性,国际上原油价格的剧烈波动必然会影响我国原油市场的价格波动。因此,研究国内外原油市场结构特征及价格波动规律对于我国原油市场的政策制定与风险管理具有重要意义.本文首先运用多重分形消除趋势波动分析法(MF-DFA),研究中石油和中石化两个上市公司股票收益率的多重分形特征,并结合多重分形谱方法,比较两股票收益率序列的多重分形性的强弱及风险大小.结果表明,两个公司的股票收益率序列均具有明显的多重分形特征,且中石化收益率序列的多重分形性更强,波动复杂程度更高.总体上相比,买入中石化股票获利的空间更大,但风险也较买入中石油的更高.其次运用交互相关统计量和多重分形分析方法研究国内和国际代表性原油市场的价格收益率序列,证实了原油市场间存在交互相关关系,且这种关系具有多重分形特征.同时,比较分析了收益率序列之间交互相关关系的多重分形性的强弱.研究了引起多重分形性的主要原因,发现小波动的交互相关性强于大波动交互相关性,小波动和大波动的长程相关性及收益率序列的胖尾分布均是形成多重分形性的主要原因.本文结构主要如下：第一章为绪论,介绍了本文的选题背景及研究意义,说明了研究国内外原油市场特征的必要性,综述了近年来国内外原油市场分形分析的一些研究成果.第二章简要介绍了MF-DFA和MF-DCCA两种多重分形分析方法.第三章首先运用MF-DFA方法,研究中石油和中石化两个上市公司股票收益率序列的多重分形性,研究表明两个公司的股票收益率序列均具有明显的多重分形特征,且中石化收益率序列的多重分形性更强,波动复杂程度更高.其次运用MF-DCCA方法研究了WTI/Brent, WTI/大庆,WTI/胜利,大庆/胜利四对原油市场的价格收益率序列,证实了原油市场间存在交互相关关系,且这种关系具有多重分形特征.同时,比较分析了收益率序列之间交互相关关系的多重分形性的强弱.分析表明市场间小波动的交互相关性是持续的,而大波动的交互相关性是反持续的.比较收益率序列的多重分形谱宽度,发现资产组合序列的金融风险测度小于自相关序列的风险测度.第四章研究WTI/Brent, WTI/大庆,WTI/胜利,大庆/胜利四对原油市场的价格收益率序列形成多重分形的成因,发现长程相关和胖尾分布均是形成多重分形的原因,并分别指明了四个原油市场之间收益率序列多重分形形成的主要原因.第五章为本文研究成果的总结与进一步研究的展望.
[Abstract]:With the increase of China's international crude oil import, the domestic crude oil price has a great dependence on the international crude oil price. The sharp fluctuation of the international crude oil price will inevitably affect the price fluctuation of our country's crude oil market. It is of great significance to study the structural characteristics and price fluctuation law of crude oil market at home and abroad for the policy formulation and risk management of China's crude oil market. MF-DFA). This paper studies the multifractal characteristics of stock returns of two listed companies in PetroChina and Sinopec, and compares the multifractal property and the risk of the two stock returns series with the multifractal spectrum method. The results show that. The stock return series of the two companies have obvious multifractal characteristics, and Sinopec yield series has stronger multifractal and more complex volatility. There is more room for profit from buying Sinopec shares. But the risk is also higher than that of buying PetroChina. Secondly, we use the interactive correlation statistics and multifractal analysis method to study the price return series of domestic and international representative crude oil market. It is proved that there is an interactive correlation between the crude oil market, and this relationship has multifractal characteristics. At the same time. In this paper, we compare and analyze the multifractal relationship between the return series, and study the main causes of multifractal. It is found that the interaction correlation of small volatility is stronger than that of large volatility. The long range correlation of small volatility and large volatility and the fat tail distribution of yield series are the main reasons for multifractal. The structure of this paper is as follows: chapter one is introduction. The background and significance of this paper are introduced, and the necessity of studying the characteristics of crude oil market at home and abroad is explained. This paper summarizes some research achievements of fractal analysis in crude oil market at home and abroad in recent years. Chapter 2 briefly introduces two multifractal analysis methods, MF-DFA and MF-DCCA. Chapter 3 uses MF-DF firstly. A method. This paper studies the multifractal character of the stock return sequence of the two listed companies of petroleum and Sinopec. The research shows that the stock return series of the two companies have obvious multifractal characteristics. And Sinopec yield sequence is more multifractal, more complex volatility. Secondly, the MF-DCCA method is used to study WTI / WTI / Daqing WTI / victory. The price return sequence of Daqing / Shengli four pairs of crude oil market confirmed the existence of interactive correlation between the crude oil market, and this relationship has multifractal characteristics. At the same time. The multifractal correlation between the return series is compared and analyzed. The results show that the interrelation between the small volatility of the market is continuous. The cross-correlation of large volatility is anti-persistent. The multifractal spectrum width of the yield series is compared. It is found that the financial risk measurement of portfolio series is smaller than that of autocorrelation series. Chapter 4th studies WTI / Brent, WTI / Daqing WTI / Victory. Four pairs of Daqing / Shengli crude oil market price yield series of the formation of multifractal causes, it is found that long-term correlation and fat tail distribution are the reasons for the formation of multifractal. The main reasons of multifractal formation of yield series between four crude oil markets are pointed out respectively. Chapter 5th is the summary of the research results and the prospect of further research.
【学位授予单位】：南京财经大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F416.22
，

本文编号：1396091