三种汇率风险暴露估计方法的比较研究

发布时间：2019-06-21 03:31

【摘要】：在人民币升值和人民币汇率波动幅度增加的背景下,我国企业和宏观经济面临日益增强的外汇风险。因此准确评估企业汇率风险的研究显得越来越重要。针对当前我国企业汇率风险暴露估计结果存在较多不一致性的问题,本文考察了目前我国学者研究中的常用方法以及国外这一领域研究方法的最新发展。从资本市场法角度,选择了最常用的和最新的三种企业汇率风险暴露估计的统计模型——Jorion模型、正交化Jorion模型以及GJR-GARCH模型。从理论和经验两个方面,比较了这三种模型在测度我国企业汇率风险暴露方面的适用性。这对于我国开展该领域的应用研究具有重要的指导意义,对于推进该领域的方法研究具有一定的理论价值。研究发现：①三种模型在测度我国企业汇率风险暴露时具有较好的一致性,但在研究精度上仍然存在差异；②Jorion模型和GJR-GARCH模型测度的是企业未预期到的汇率风险暴露。与其它两种模型不同,正交化Jorion模型的估计结果同时包含了直接汇率风险暴露和间接汇率风险暴露,这使得该方法的经济学意义与其它两种模型存在差异；③检验结果表明,我国金融市场数据具有尖峰厚尾性、异方差性和非对称波动性等特征,因此在选择测度模型时需要考虑上述数据特征,从理论上看GJR-GARCH模型是最适宜测度我国企业汇率风险暴露的模型。④实证分析证明,在三种模型中,GJR-GARCH模型具有最小的拟合残差,拟合优度最高。从拟合结果看GJR-GARCH模型是最适宜测度我国企业汇率风险暴露的模型。
[Abstract]:Under the background of RMB appreciation and RMB exchange rate fluctuation, Chinese enterprises and macroeconomy are facing increasing foreign exchange risk. Therefore, it is more and more important to accurately evaluate the exchange rate risk of enterprises. In view of the fact that there are many inconsistency in the estimation results of exchange rate risk exposure of Chinese enterprises, this paper investigates the common methods in the research of Chinese scholars and the latest development of foreign research methods in this field. From the point of view of capital market method, three kinds of statistical models, Jorion model, orthogonal Jorion model and GJR-GARCH model, which are the most commonly used and the latest statistical models for enterprise exchange rate risk exposure estimation, are selected. This paper compares the applicability of these three models in measuring the exposure of exchange rate risk of Chinese enterprises from two aspects of theory and experience. This has important guiding significance for the application research in this field in our country, and has certain theoretical value for promoting the method research in this field. It is found that: (1) the three models have good consistency in measuring the exchange rate risk exposure of Chinese enterprises, but there are still differences in the research accuracy, and the 2Jorion model and GJR-GARCH model measure the unexpected exchange rate risk exposure of enterprises. Different from the other two models, the estimation results of orthogonal Jorion model include both direct exchange rate risk exposure and indirect exchange rate risk exposure, which makes the economic significance of this method different from the other two models. The test results show that the financial market data in China have the characteristics of sharp peak and thick tail, heteroscedasticity and asymmetric volatility, so it is necessary to consider the above data characteristics when selecting the measurement model. Theoretically, GJR-GARCH model is the most suitable model to measure the exchange rate risk exposure of Chinese enterprises. 4 empirical analysis shows that among the three models, GJR-GARCH model has the smallest fitting residual and the highest goodness of fit. From the fitting results, the GJR-GARCH model is the most suitable model to measure the exchange rate risk exposure of Chinese enterprises.
【学位授予单位】：天津财经大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F224;F832.6

【参考文献】