【作者】 卢浩；

【导师】 黄运成；

【作者基本信息】 中国科学技术大学 ， 金融工程， 2010， 博士

【摘要】 权证是一种赋予持有人在约定时间内有权按约定价格向发行人购入或者出售合同规定的标的证券的凭证,是成熟证券市场比较流行的金融工具。随着全球范围内权证市场的兴起,作为一种重要的衍生工具,针对权证的定价、风险管理以及市场微观结构的研究日渐成为关注焦点。市场的微观结构是从微观的角度去研究金融市场的特性和内在规律,已成为当今金融研究领域的热点之一。波动性和流动性作为金融市场最重要的特性,已成为反映市场质量和效率的最重要的指标,并始终贯穿整个市场微观结构理论。针对市场微观结构的研究,依靠低频交易数据已经不能满足研究的需要。伴随着现代计算机技术的发展,高频数据逐渐被研究者们应用到对市场微观结构的研究中。实证研究表明,利用高频数据针对市场的波动性和流动性展开建模分析具有优良的特性。本论文首先介绍了中国权证市场发展状况和金融高频数据的研究进展,接着回顾了微观结构的研究内容和国内外的研究现状,随后分别就波动性和流动性展开理论探索和实证研究,得出结论。针对波动性,本论文首先介绍了波动性的定义、特征和估计模型,然后选取中国权证市场的13只权证,利用分时高频交易数据,提取价格收益率序列,建立了相应的GARCH模型,随后,又结合IGARCH和TGARCH模型针对市场上存在的不对称效应进行了分析。针对流动性,首先介绍了流动性的概念、度量以及ACD估计模型,在此基础上,利用权证的分笔高频交易数据,选取交易量持续期这一指标来刻画流动性,经过数据处理和交易量持续期的提取,对模型的有效性、分布假设进行了充分的研究和探讨。根据研究结果可以判断,ACD模型在分析权证市场的微观结构方面具有非常好的表现。进一步我们将新的微观结构变量加入原先模型中,构造了以检验市场微观结构假设的扩展ACD模型,并对市场微观结构理论的几个重要假设进行了实证检验,检验表明平均交易量和买卖价差因子对市场流动性具有显著负面效应。本文的研究表明：GARCH模型及其扩张的IGARCH模型可以很好地模拟权证市场的波动性,证明了权证市场的波动存在明显的集聚性,并且这种波动在一定程度上是非对称的。ACD模型和SCD模型可以很好地模拟权证市场的流动性。模型拟合结果表明权证市场的交易量持续期存在明显的集聚现象,ACD和SCD模型可以很好地对此予以解释。综合考虑模型估计和残差检验,对数GACD模型和SCD模型是流动性拟合效果最好的两种模型,根据拟合出的模型就可以对权证的波动性和流动性进行预测。根据研究结果,我们有针对性地提出了一些政策建议,通过改革交易机制、设定创设资产规模限制、发展权证新品种等措施,可以在一定程度上减少波动集聚和提升流动性,从而实现权证市场的健康发展。更多还原

【Abstract】 A warrant is a security that entitles the holder to buy stock of a company that issued at a specified price within the certain time, it’s one of the most popular financial instruments in developed security market. As the market developing, the research on the pricing, risk management and microstructure of warrant has become the focus of researchers.Market microstructure is about research on the features and patterns of financial market, which is becoming the hotspot of finacial fields. Volatility and liquidity have been the most important indicators of market quality and efficiency as they are the most important features of financial market.Researches in the market microstructure focus on the processes of discovering the trading prices and the operating mechanisms of the trading prices. Low frequency trading data is far from enough for this kind of researches. With the rapid development of computer science, it makes possible for people to explore the market microstructure with high frequency data. The empirical researches indicate that the model constructed for volatility and liquidity exhibits superior features in dealing with the high frequency financial data.In this paper, we first introduce the development of the warrant market of Chinese mainland and the application of high frequency financial data, then give a review of the research on the market microstructure, after that, we conduct empirical researches for volatility and liquidity.For volatility research, we first introduce the definiton and features of volatility, then with the time sharing high frequency data of 13 warrants of mainland market, we build the GARCH model by price return series. Besides, we analyze the asymmetric effect with the combination of IGARCH and TGARCH model.For liquidity research, we first take a brief review of the related literature about liquidity. Then we introduce econometric models used in the research including the linear and nonlinear model with different distributions. As a consequence, the paper constructs and analyzes the duration of 13 selected warrants from Chinese mainland warrant market with the ACD model introduced. Then, the empirical research is conducted to identify the reasonable distribution of the model. The empirical result indicates that the ACD model does a good job in analyzing the high frequency data of warrants. With external variables of market microstructure and an extended ACD model, the empirical research confirms some important hypothesis of the microstructure of Chinese financial market.We have found from the results that the GARCH model and the extended IGARCH fit the volatility of warrant well, which prove the existence of volatility clustering with asymmetric effect. Besides, the ACD model and SCD model have a good fitting for the liquidity of Chinese mainland warrant market. The model can also explain the volatility clustering of trading volume duration exactly. Among all the models, Log-GACD model and SCD model are the most fitted if we consider the results of model estimation and residual test. We give some advice for warrant market supervisor based on the research result. By trade mechanism reform, restriction on the asset size, development of new class warrant, we can reduce the volatility clustering and improve the market liquidity, then achieve the stable development of warrant market.更多还原