基于GARCH-CVaR模型的人民币汇率风险测度研究

发布时间：2018-05-23 19:23

本文选题：人民币兑美元汇率 + CVaR　；参考：《成都理工大学》2017年硕士论文

【摘要】：2014年5月以来,美元对全球货币走强,进入长期升值通道,在此新背景下,人民币相对于美元进入贬值周期。在当今错综复杂的国际经济背景下,中国经济结构正在转型,从以前国际贸易顺差过大、外汇储备高速增长的旧常态转向国际收支趋于平衡、外汇储备数量逐渐减少的新常态。国际外汇市场汇率变动起伏不定,国与国之间汇率博弈变得更加频繁,这就使得涉外经济主体处于一种更加严峻的汇率风险环境中,汇率市场的剧烈波动势必会对其造成很大的影响,这就需要我们对新形势下的汇率市场风险度量进行更深一步研究。国内外学者已经从多个方面对关于金融市场风险的相关领域进行了深层次的研究,建立了多种能够测度全球金融市场风险的研究方法,并将其应用到股票、基金、债券等市场风险度量当中进行实证研究,取得了非常好的效果。本文选用了目前国际上最流行的风险测度方法VaR,这也是全球各大金融机构采用的风险管理新标准。尽管VaR风险测度方法本身存在着一些缺陷,不满足一致性公理以及不能预测极端风险的情况,但是不妨碍本文的应用。为了弥补这些不足,国外学者对VaR风险测度方法进行了扩展,提出了CVaR方法来预测极端情况下的风险发生,同时也满足一致性公理,该方法已经在很多研究领域得到了广泛应用,证实了方法的实用性,因此本文也采用了CVaR方法对人民币汇率风险进行度量,并与VaR风险测度方法计算结果进行比较。本文以人民币兑美元汇率作为研究对象,建立人民币兑美元汇率的对数收益率时间序列,运用GARCH和EGARCH两个模型来对汇率波动性进行刻画,并假设条件收益率分别服从正态、学生t、GED分布进行动态风险价值VaR和CVaR测度。最终结果表明在对人民币兑美元的汇率市场上使用基于EGARCH模型的GED分布下的CVaR方法进行风险价值测度可以有效刻画极端风险。
[Abstract]:Since May 2014, the dollar has strengthened to the global currency and entered a long-term appreciation channel. Under this new background, the RMB is entering a devaluation period relative to the dollar. Under the complex international economic background, China's economic structure is transforming, from the former international trade surplus, the old normal state of the foreign exchange reserve to the international balance of payments. The exchange rate fluctuations in the international foreign exchange market are fluctuating, and the exchange rate game between countries and countries becomes more frequent. This makes the foreign economy subject in a more severe exchange rate risk environment, which is bound to have a great impact on the exchange rate market, which is necessary. We should make a deeper study of the risk measurement of the exchange rate market under the new situation. Scholars at home and abroad have made a deep study on the related fields of financial market risk from many aspects, and have established a variety of research methods that can measure the risks of global financial markets, and apply them to the market winds of stocks, funds and bonds. This paper selects the most popular risk measurement method VaR in the world, which is also the new risk management standard adopted by the world's major financial institutions. Although the VaR risk measurement method itself has some defects, it does not meet the conformance axiom and can not predict the extreme risk. In order to make up for these shortcomings, in order to make up for these shortcomings, foreign scholars have extended the risk measurement method of VaR, proposed CVaR method to predict the occurrence of risk in extreme cases, and also meet the consistency axiom. This method has been widely used in many research fields, and proved the practicability of the method. This paper also uses the CVaR method to measure the RMB exchange rate risk and compares it with the results of the VaR risk measurement method. In this paper, the RMB / US dollar exchange rate is used as the research object to establish the time series of the RMB's exchange rate on the US dollar exchange rate, and use the two models of GARCH and EGARCH to describe the exchange rate volatility. And assuming that the rate of conditional return obeys normal, the dynamic value VaR and CVaR measure of Student t and GED distribution are measured. The final result shows that using the CVaR method based on the GED distribution based on EGARCH model in the exchange rate market of the RMB against US dollar can effectively depict the extreme risk.
【学位授予单位】：成都理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F832.6

【参考文献】