国内外石油价格收益率波动性分析

发布时间：2018-12-07 18:22

【摘要】：石油是目前世界上最重要的一种化工原料和战略资源，充斥着生产生活各个领域，对于各国经济的发展都起着至关重要的作用。然而，随着现代工业的迅速发展，世界各国又都面临着不同程度的能源紧缺问题。因此，石油不仅仅是一般意义上的商品，更具有政治属性，对全球政治格局、经济地位和军事形势都起着决定性的作用。同时，随着各国金融市场的迅速发展，各种衍生品的推出，石油市场与金融市场的联系日渐紧密，又呈现出一定的金融属性。作为同时兼具商品属性、政治属性与金融属性的特殊的商品，石油价格受到多种因素的影响，表现出很大的波动性。我国作为目前世界上最大的石油进口国之一，石油进口依存度很大，但对于油价制定的话语权却非常有限。国际石油价格的剧烈波动势必会传至我国，从而对我国经济造成重大影响。因此，研究国内外石油市场价格的波动性规律，对我国石油市场的风险监管具有重要的意义。波动性是指价格非预期变化的趋势或收益的概率分布，在早期以有效市场假说为基础的金融研究中，波动性通常被假设为某种固定形式（如收益率方差固定不变或取平均值）。但后来很多学者研究发现，金融资产收益率的波动，在时间序列上表现出尖峰厚尾、集簇等特征。为此，学者引入不同的波动模型进行分析，目前主要分为GARCH和随机波动两类模型。本文采用了定性分析和实证研究相结合的方法，分析了国内外石油价格波动的特征。全文共分五章进行阐述：第一章是绪论，对石油价格波动的研究背景、意义、国内外研究现状、全文的结构框架及创新点进行了介绍。第二章是全文的理论基础，介绍了金融市场的波动性特征以及文章实证部分所采用的主要模型和一些相关的方法理论。第三章对国内外石油市场的波动性特征作了定性研究，主要包括石油的战略地位、石油的战略地位，以及后危机时代石油价格波动的具体情况。第四章是全文的重点，利用GARCH和SV模型族刻画了WTI与大庆石油价格波动的特征，并且利用样本外预测的方法对两类模型的预测能力进行了比较。最后，第五章为全文的结论。本文的主要观点是：第一，，国内外石油价格波动均存在一定的集聚性和持续性，通过GARCH模型建模，能较好地消除ARCH效应。并且，误差分布假设为t分布时的模拟结果明显优于另外两种分布假设。第二，国内外石油市场均存在明显的杠杆效应，加入对杠杆效应的刻画，能提高模型对石油价格波动性预测的能力。第三，国内外石油市场均呈现出“高风险、高收益”的特征，并且，SV模型类对该特征的刻画能力优于GARCH-M模型。第四，综合比较，在刻画石油价格收益率波动性方面，随机波动模型优于GARCH模型族。
[Abstract]:Oil is one of the most important chemical raw materials and strategic resources in the world, which is full of production and life, and plays a vital role in the economic development of all countries. However, with the rapid development of modern industry, all countries in the world are facing the problem of energy shortage to varying degrees. Therefore, oil is not only a commodity in general sense, but also a political attribute, which plays a decisive role in the global political structure, economic status and military situation. At the same time, with the rapid development of financial markets in various countries and the introduction of various derivatives, the oil market and the financial market are increasingly closely linked, showing a certain financial attributes. As a special commodity with both commodity attributes, political attributes and financial attributes, the oil price is affected by many factors, showing great volatility. As one of the largest oil importing countries in the world, China has a great dependence on oil import, but it has a very limited right to say about oil price. The sharp fluctuation of international oil price is bound to spread to China, which has a great impact on our economy. Therefore, it is of great significance to study the volatility of oil market price at home and abroad. Volatility refers to the trend or probability distribution of the unexpected change of price. In the early financial research based on the efficient market hypothesis, volatility is usually assumed to be a certain fixed form (such as fixed or average return variance). However, many scholars later found that the volatility of financial asset return shows the characteristics of peak, thick tail and cluster in time series. For this reason, scholars introduce different volatility models, which are mainly divided into two types: GARCH model and stochastic volatility model. In this paper, the characteristics of oil price fluctuation at home and abroad are analyzed by combining qualitative analysis with empirical research. The paper is divided into five chapters: the first chapter is the introduction, the research background, significance, domestic and foreign research status, the structure of the full text and innovation points are introduced. The second chapter is the theoretical basis of the full text, introduced the volatility characteristics of financial markets, the empirical part of the main models and some related methods. The third chapter makes a qualitative study on the volatility characteristics of oil market at home and abroad, mainly including the strategic position of oil and the specific situation of oil price fluctuation in the post-crisis era. The fourth chapter is the focus of this paper. The characteristics of WTI and Daqing oil price fluctuation are described by using GARCH and SV model families, and the prediction ability of the two models is compared by using the method of extrasample prediction. Finally, the fifth chapter is the conclusion of the full text. The main points of this paper are as follows: first, there is a certain concentration and persistence of oil price fluctuation at home and abroad. The ARCH effect can be eliminated by GARCH model modeling. Moreover, when the error distribution is assumed to be t distribution, the simulation results are obviously superior to the other two distribution assumptions. Secondly, there is obvious leverage effect in both domestic and foreign oil markets. The ability of model to predict the volatility of oil price can be improved by adding the depiction of leverage effect. Thirdly, the oil market at home and abroad presents the feature of "high risk and high profit", and the SV model class has better characterizing ability than the GARCH-M model. Fourthly, the stochastic volatility model is superior to the GARCH model in describing the volatility of oil price return.
【学位授予单位】：浙江财经大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F416.22;F764.1

【参考文献】