基于简化WCVaR模型的金融极端风险度量研究

发布时间：2018-01-10 14:06

本文关键词：基于简化WCVaR模型的金融极端风险度量研究　出处：《合肥工业大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着金融自由化和全球化的不断发展,国际金融市场的交易规模也在不断壮大。进入二十一世纪以来,计算机信息技术的发展使得金融产品和金融技术革新的速度达到前所未有的高度。然而,由于资本的逐利性,技术的发展并不能改变金融行业的本质,反而使金融行业之间的竞争愈演愈烈。为了取得超额收益,交易员不惜铤而走险。各种金融产品的发明给投资者提供便利的同时,也加大了整个市场的投机性。整个金融市场的波动性越来越大,充斥着各种不稳定因素。国内方面,大连证券、新华证券先后倒闭,富友证券、南方证券被政府托管。国际方面,从英国巴林银行、美国雷曼兄弟倒闭,到如今的欧洲国家主权债务危机,上述机构之所以走向危机,究其原因,均为其交易员或者管理层对极端风险重视不足,最终导致严重的后果。因此,金融风险管理已经成为现代金融业的核心之一。因为金融市场具有复杂性和尖峰肥尾性,目前常用的风险度量方法对分位点下方的风险重视不够,现有方法已不太适用当今的金融环境。事实上,由于“黑天鹅”事件预测和控制难度大,破坏性和冲击性往往更是灾难性的。本文围绕这一主题,主要进行以下的研究:(1)金融极端风险的理论研究。面对日益复杂的国内外金融环境,传统的金融风险度量方法对极端环境下的风险往往重视不够。本文主要在目前常用的金融风险度量工具的基础上进行延伸并加以改进,提出了简化的WCVaR模型,并对其进行理论推导。(2)金融极端风险的实证分析。本文选用国内外金融市场的主流指数进行实证分析,分别通过传统的VaR法、CVaR法和简化的WCVaR方法对金融尾部风险进行度量。实证分析主要分为两个部分,第一部分基于历史模拟法对以上方法进行求解以验证该方法的可行性;第二部分对三种风险度量方法进行事后检验,对比不同方法对金融风险度量的有效程度。(3)基于简化的WCVaR模型的投资组合优化管理。本文在原有风险度量方法的基础上提出了简化的WCVaR风险度量模型。由于该模型具有高维复杂的max-min结构,普通的线性规划算法很难求解。本文首次采用粒子群优化算法对其进行求解,为极端条件下的风险度量和风险规避探索出一种新的思路。最后通过实证,验证了WCVaR模型在风险度量和资产风险规避方面具有一定的应用价值。
[Abstract]:With the continuous development of financial liberalization and globalization, the scale of international financial markets is also growing. Since 21th century. With the development of computer information technology, the speed of financial products and financial technology innovation has reached an unprecedented high. However, because of the profit-driven nature of capital, the development of technology can not change the nature of the financial industry. Instead, competition between the financial industry intensified. In order to achieve excess returns, traders did not hesitate to take risks. The invention of various financial products provides convenience to investors at the same time. It also increases the speculative nature of the whole market. The whole financial market is becoming more and more volatile, full of various unstable factors. At home, Dalian Securities, Xinhua Securities went bankrupt one after another, Fuyou Securities. Southern Securities was entrusted by the government. Internationally, from the British Bank of Bahrain, the collapse of Lehman Brothers, to the current European sovereign debt crisis, the reasons for the crisis are investigated. For their traders or management to pay less attention to extreme risks, and ultimately lead to serious consequences. Financial risk management has become one of the core of modern financial industry. Because of the complexity and sharp tail of financial market, the current risk measurement methods do not pay enough attention to the risk under the sub-locus. The existing methods are not suitable for the present financial environment. In fact, because of the difficulty of prediction and control of "Black Swan" events, the destructive and impact is often more catastrophic. This paper focuses on this theme. Mainly carry on the following research: 1) the financial extreme risk theory research. Facing the increasingly complex domestic and foreign financial environment. The traditional financial risk measurement methods often pay less attention to the risk in the extreme environment. This paper mainly extends and improves the financial risk measurement tools which are commonly used at present. A simplified WCVaR model is put forward, and the empirical analysis of the extreme financial risk is carried out. In this paper, the mainstream index of the domestic and foreign financial markets is selected for empirical analysis. Through the traditional VaR method Cvar method and the simplified WCVaR method to measure the financial tail risk. The empirical analysis is divided into two parts. The first part is based on the historical simulation method to solve the above method to verify the feasibility of the method; The second part of the three risk measurement methods after the test. Compare the effectiveness of different methods to measure financial risk. Portfolio optimization management based on simplified WCVaR model. This paper proposes a simplified WCVaR risk measurement model based on the original risk measurement method. The model has high dimension and complex max. -min structure. The ordinary linear programming algorithm is difficult to solve. Particle swarm optimization algorithm is used to solve the problem for the first time in order to find a new way of thinking for risk measurement and risk avoidance under extreme conditions. The application value of WCVaR model in risk measurement and asset risk aversion is verified.
【学位授予单位】：合肥工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F831

【相似文献】