黄金价格收益率的杠杆效应研究

发布时间：2018-01-03 16:36

本文关键词：黄金价格收益率的杠杆效应研究　出处：《西南财经大学》2014年硕士论文　论文类型：学位论文

【摘要】：黄金最初作为支付手段,充当世界范围内的货币。而1971年黄金与美元脱钩以后,黄金开始向贮藏和保值的职能方向转化。自亚洲金融危机和2008年美国金融危机后,美元的贬值和世界范围内经济形势的不确定因素的增加,黄金需求结构受到很大的影响和变化,需求主力开始转向投资领域。黄金作为一种新兴的投资工具开始形成其特有的市场。黄金作为投资组合中的一种重要资产,在保值增值和风险规避方面发挥着不可替代的作用。经济形势决定黄金投资的地位,在不确定的经济环境中,投资黄金已经成为了继证券、期货、外汇之外,独具吸引的一种投资渠道。从而投资者开始关注影响黄金价格的因素,以及黄金价格收益率具有的特征。黄金价格受多种因素的影响,比如美元,股票指数,通货膨胀指数等,而这些影响因素都得到了很多学者的研究,且已经得出很多一致和成熟的结论。黄金作为金融市场的投资品,其价格收益率的波动性是否也存在一定的特征?分析并掌握该种特征对黄金市场投资是否能起到一定的指导作用?如果黄金市场价格收益率波动性存在自身的特征,能否给投资者一些投资建议? 一般金融资产收益率的波动率普遍存在一些共同的特征。首先,波动率普遍存在波动率聚集。其次,波动率以连续的方式随时间变化。再次,波动率是一个平稳的时间序列。最后,波动率对价格的大幅上升和大幅下降存在着不同的反应,也即不同性质信息的冲击对波动率的影响程度存在着不对称的效应。这种效应普遍称为杠杆效应。一般金融资产价格收益率的波动率普遍存在前三种特征,但杠杆效应在有些市场上存在,而在另一些市场上却不存在。黄金市场是否存在这样的杠杆效应,不同的文献得出不同的结论。所以,本文主要针对中美黄金市场是否存在杠杆效应展开了研究。专门选择中国和美国黄金市场主要基于以下原因。首先中国作为黄金消费的第一大国,其不断增长的黄金消费量将会引起金商们投资黄金的持续动机,且中国近来高企的CPI也使得个人开始以投资黄金的方式规避货币贬值、通货膨胀。其次由于美国经济发展形势和美元地位对世界经济的影响力较大,同时美国作为世界黄金期货交易量最大的市场,所以在黄金定价话语权上,美国黄金市场是世界黄金价格的风向标。针对黄金市场杠杆效应的研究内容,本文确定了以下样本和研究方法。首先选择的样本是中国和美国的黄金市场的现货和期货从2008年到2013年的数据。其次方法采用了研究杠杆效应的三种GARCH模型,即TGARCH,EGARCH和PGARCH。在运用上述三种模型时,对均值方程和残差分布进行了一定的修正。在文章结构上,本文主要分为五个章节。第一章为引言部分,主要是研究背景、研究意义、文章的研究思路和基本框架以及本文的特点和不足。第二章为文献综述部分,主要对国内和国外的文献在黄金市场上的研究进行整体的梳理。本章包括两部分,1、黄金价格的影响因素；2、杠杆效应研究及黄金市场存在的杠杆效应特征。首先第一部分主要是为实证模型中外生变量的选取提供理论基础。其次第二部分杠杆效应的相关文献总结为黄金市场的杠杆效应提供了具体表现形式的描述。第三章主要介绍黄金的固有属性及其地位的历史演变,黄金市场的参与者和组织结构,世界主要的黄金市场概述,中国黄金市场现状,以及影响黄金的主要因素分析。对黄金价格影响因素的分析得出,黄金价格在长期范围内主要受美元、主要国家的股票指数、商品价格指数和原油价格等的影响。第四章是本文的实证部分。本章分为六个小节。第一节对黄金价格的收益率的研究意义进行了描述。第二节对一般研究收益率的模型进行了介绍。第三节是对实证模型的选择。本文主要是针对黄金价格收益率波动性在杠杆效应方面的研究,而研究杠杆效应的模型有TGARCH,EGARCH和PGARCH三种,所以本文选择了这三个模型研究黄金市场的杠杆效应。第四节是实证模型的修正,即均值方程加入了外生变量和风险溢价因素的影响。第五节是实证数据的选择。第六节是实证过程。实证过程中运用修正后的研究杠杆效应的三种GARCH模型进行数据拟合,并通过AIC准则选择最佳拟合模型。第五章根据中美黄金市场的实证结果得出本文的结论,并分析两个市场在杠杆效应方面存在差异的原因。此外还包括相关投资建议和进一步研究方向。经过实证分析,本文得出以下结论。 1、中国黄金市场无论是现货还是期货,均不存在针对利好信息和利空信息的杠杆效应。美国黄金现货价格收益率波动率存在杠杆效应,而黄金期货价格收益率波动性不存在杠杆效应。 2、美国黄金现货市场存在的杠杆效应情况与一般研究的杠杆效应不同。一般金融资产存在的杠杆效应是对利空信息反应程度大于对利好信息的反应程度,而本文得出的美国黄金现货价格收益率波动存在的杠杆效应正好与此相反,即其对利好信息的反应程度大于对利空信息的反应程度。本文存在如下特点。 1、在内容上主要研究中、美黄金市场是否存在杠杆效应,以及在杠杆效应的差异性和形成原因上进行分析和解释。 2、在研究模型上,选择了研究杠杆效应的三种GARCH模型,即TGARCH, EGARCH和PGARCH,并对三种模型进行了对比分析。 3、在模型的修正上,将美元指数、商品价格指数、股票指数和黄金价格收益率波动率的标准差分别作为外生变量和风险溢价因素同时加入到均值方程中。文章后续的研究方向有,1、分时期进行黄金市场现货和期货杠杆效应研究,并对比和分析各个时期的结果；2、本文选择的黄金市场仅有中国和美国市场,将全球主要黄金市场均纳入研究范围也是后续研究方向；3、在杠杆效应研究方面,我们假设的利好利空信息均是对未来的预期,该类信息并没有发生,在模型中也仅是一个假设。而已经发生的利好利空信息对市场仍然有着深远的影响。该影响可能与预期信息共同对黄金市场存在的杠杆效应产生影响,所以,针对已有信息对杠杆效应的反作用,也是进一步研究的方向。
[Abstract]:The original gold as a means of payment, as the world currency. And in 1971, gold and the dollar after decoupling, gold began to transform to the direction of the storage and preservation of function. Since the Asian financial crisis and the United States in 2008 after the financial crisis, the depreciation of the dollar and the world economic situation of uncertainty increases, gold demand structure by great changes and influence of the main demand to gold investment. As a new investment tool began to form its unique gold market. As an important asset in the portfolio, plays an irreplaceable role in increasing the value and risk aversion. The economic situation determines the status of investment in gold. The uncertain economic environment, investment gold has become the following securities, futures, foreign exchange, a unique investment channels. In order to attract investors began to focus on the effect of yellow The gold price factors and the characteristics of the gold price, rate of return is. Gold price is affected by many factors, such as the dollar, stock index, inflation index, but these factors have been studied by many scholars, and has obtained many favorable and mature conclusion. Gold as a financial market investment goods, the price the volatility of whether there is a certain feature? Analyze and grasp this feature of investment gold market can play a guiding role? If the gold market price return volatility has its own characteristics, can give investors some investment advice?
The general rate of financial asset volatility generally have some common characteristics. First, the volatility of common volatility aggregation. Secondly, the volatility in the continuous mode changes with time. Again, the volatility is a stationary time series. Finally, the volatility of large amplitude of rising prices fell sharply and there different reactions, namely the different nature of the impact of information asymmetry exists the effect on volatility. This effect is commonly referred to as "leverage effect. Generally the price of financial assets yield volatility exists before the three kinds of features, but the leverage effect exists in some markets, but does not exist in other markets. Whether there is leverage effect that the gold market, the different literature come to different conclusions. Therefore, this paper focuses on the Sino US the existence of the gold market leverage effect is studied. Specially selected China and the gold market is mainly based on the following reasons. Firstly, Chinese as the largest consumption of gold, gold consumption will continue its growing motivation of gold producers to invest in gold, and Chinese recent high CPI also makes personal investment gold party began to avoid the currency devaluation, inflation. Secondly due to the situation of America the development of economy and the dollar's influence on the world economy is larger, while the United States as the world's largest gold futures market, so the right to speak in the gold price, the United States is the world gold market gold price benchmark.
The research content for leverage effect of the gold market, this paper identified the following samples and research methods. The sample data is the first choice of the gold market and Chinese spot and futures from 2008 to 2013. The second method uses three kinds of GARCH models, the leverage effect of TGARCH, EGARCH and PGARCH. in the use of the above three model, the mean equation and residual distribution is modified.
In the structure of the article, this article is divided into five chapters.
The first chapter is the introduction, which is mainly the research background, the research significance, the research ideas and the basic framework of the article, and the characteristics and shortcomings of this article.
The second part is literature review, research in the gold market mainly on domestic and foreign literatures overall combing. This chapter includes two parts, 1 factors that affect the price of gold; 2, leverage characteristics have lever effect research and the gold market. The first part is mainly to provide the theoretical basis for the selection of an empirical model of Chinese and foreign students. The variable leverage effect literature followed by the second part of the leverage effect summary for the gold market provides specific forms of description.
The third chapter mainly introduces the historical evolution of the inherent attributes of gold and the status of the participants and the organizational structure of the gold market, the world's major gold market, gold market situation China, analysis and main factors affecting the price of gold gold. The influence factors analysis, the dollar price of gold mainly in long range, major countries the stock index, the impact of commodity price index and the price of crude oil and so on.
The fourth chapter is the empirical part of this paper. This chapter is divided into six sections. The first section on the price of gold yields the significance of the research are described. The second section of the general research yield model are introduced. The third section is the empirical model choice. This article is mainly focused on the gold price volatility. The leverage effect, and the leverage effect of the model with TGARCH, EGARCH and PGARCH three, so this paper chooses the leverage effect of the three models of the gold market. The fourth section is the empirical correction model, namely the mean equation joined the influence factors of exogenous variables and risk premium. The fifth section is the empirical data. Selection. The sixth section is the empirical process. Three kinds of GARCH model with leverage effect of the modified empirical process of data fitting, and by the AIC criterion to choose the best fitting model.
The fifth chapter draws the conclusion based on the empirical results of the Sino US gold market, and analyzes the reasons for the differences between the two markets in terms of leverage effect. Besides, it also includes relevant investment suggestions and further research directions.
After the empirical analysis, this paper draws the following conclusions.
1, there is no leverage effect in China's gold market, no matter spot or futures, which is good for information and bad information. There is leverage effect in the US gold spot price volatility, and there is no leverage effect in the volatility of gold futures price returns.
2, the situation of leverage effect exist in the United States gold spot market and the leverage effect of general studies. The leverage effect exists in financial assets is generally bad information is greater than the degree of reaction degree of positive reaction to the information, and the American gold spot price return volatility leverage effect exists for the opposite, namely the reaction the degree of positive information is greater than the degree of response to bad news.
This article has the following characteristics.
1, in the main content of the study, whether there is a leverage effect in the US gold market, and to analyze and explain the differences and causes of the leverage effect.
2, on the research model, three kinds of GARCH models, namely, TGARCH, EGARCH and PGARCH, are selected to study the leverage effect, and the three models are compared and analyzed.
3, in the revision of the model, the standard deviations of the US dollar index, commodity price index, stock index and gold price return volatility are taken as exogenous variables and risk premium factors simultaneously, and added to the mean value equation.
The research direction in the later period, 1 points, gold spot and Futures Market Research and leverage effect, comparison and analysis of the results of each period; 2, the only Chinese gold market and the U.S. market, the global major gold markets are also included in the scope of the study is the follow-up research direction; 3, study on the lever we assume that the effect, good bad information is expected for the future, this kind of information does not occur in the model, it is only a hypothesis. And have been good bad information still has a profound impact on the market. This effect may be related to a common impact on the expected information, the gold market has leverage effect so, according to the existing information on the role of anti leverage effect, but also the direction for further research.

【学位授予单位】：西南财经大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F224;F832.54

【参考文献】