【作者】 杨琪；

【导师】 张筱峰；

【作者基本信息】 长沙理工大学 ， 金融学， 2011， 硕士

【摘要】 金融资产收益的波动率研究是当代金融经济学和计量经济学研究的核心领域,作为市场风险的度量,对波动率的辨识将直接影响到资产定价、资源配置以及风险管理的建模。一方面,波动率与市场的不确定性和风险直接相关,波动性增加,表示风险的增加,高风险将对企业经营造成的冲击较大,而收益—风险的均衡是资本配置与资产定价的核心;另一方面,波动率与反映股票市场质量和效率的其他指标密切相关,这些指标包括流动性、透明性、交易成本、市场效率、信息流动特性等,它们是最简洁和最有效的综合反映股市价格行为、质量和效率的指标之一。除此之外,波动率对企业财务杠杆、营运杠杆和投资杠杆的决策,对个人的消费与投资行为,以及对经济周期和相关宏观经济变量等都有着不可低估的作用。正因为如此重要,波动率成为金融市场不变的主题。本文针对我国货币市场基准利率的波动进行研究,选取上海银行间同业拆放利率的隔夜、一周、三月及一年利率日数据作为研究样本。首先介绍了波动理论及各种波动估计模型,主要介绍了GARCH族模型。接着分析了选取利率序列的统计特征。从统计特征图和QQ图可看出,SHIBOR市场各期限利率序列不服从正态分布,短期序列具有尖峰厚尾特征;长期序列一阶平稳,无ARCH效应。然后借助五种GARCH类模型对上海银行间同业拆放短期利率各个方面的波动特征进行实证研究,得出模拟SHIBOR隔夜利率序列及一周利率序列波动最佳的模型。实证结果显示,在GARCH族模型(EGARCH除外)拟合下,本文所选取利率序列不存在条件异方差和残差序列相关现象,模型的拟合效果很好。基于正态分布的SHIBOR利率序列方差方程存在明显的非对称效应。从文中采用的模型来看,根据AIC和SC信息准则,无论是SHIBOR隔夜利率序列还是一周利率序列,PGARCH拟合下的方差方程都比其他模型的方差方程的拟合效果要好。最后对本文进行了总结与展望,为今后进一步深化对我国货币市场风险特征的研究提供一定的参考。更多还原

【Abstract】 Analysis on the volatility of financial assets is the nucleus realm of the research of contemporary financial economics and econometrics. As the generous character of market risk, the identification of volatility will have a direct impact on assets pricing, resource arrangement and the model of venture management. On one hand, the volatility is directly related to uncertainties and risks of the market, and the core of asset pricing and capital configuration is the balance between benefits and risk; on the other hand, it closely related to other indicators which reflect the quality and efficiency of the stock market such as liquidity, market efficiency, transaction costs, transparency and information flowing characteristics, is as one of the most simple and effective indicator that can comprehensively reflect the price behavior, quality and efficiency of stock market. In addition, the volatility has an important influence on the decision to business investment leverage, financial leverage and operating leverage, on personal consumption and investment behavior, and on the economic cycle and the relevant macroeconomic variables. Because of its importance, volatility becomes a constant theme of financial market.This paper is a research on the volatility of benchmark interest rate in currency market, selecting the data of SHIBOR for overnight, one-week, one-season and one-year interest rate as the study sample. Firstly the author introduces the wave theory and a variety of volatility estimation model, main part is the GARCH models. Then author analyzes the statistical characteristics of selected sequences of interest. Seen from the statistical characteristical maps and QQ plans, SHIBOR term market interest rate sequence does not obey the normal distribution, bringing about a short sequence with a fat tail and long sequence of first-order stationary, no ARCH effects. And then various aspects in short-term interest rate characteristics have empirically researched with five GARCH type models, obtained the best model on simulation SHIBOR overnight rate sequences and sequences of the week fluctuations in interest rates. The empirical results show that in the GARCH models (EGARCH excluded), sequence of interest rates which the paper selected does not exist phenomenon of the conditional heteroskedasticity and residual sequence, actually the model fit very well. SHIBOR Interest rate sequence variance equation based on normal distribution exist obvious asymmetric effect. From the model used in the text view, according to AIC and SC information criteria, whether it is SHIBOR interest rates overnight rate series or sequence of the week, the variance equation under PGARCH fits better than that of other models . Finally, this paper make a summary and give prospects for future, providing some reference for further study of deepening the money market risk characteristics.更多还原