基于高频数据的量价动态关系研究

发布时间：2018-08-23 12:28

【摘要】：近年来随着计算技术与电子交易系统的不断发展,以及交易成本的下降,金融市场中数据的获取及处理方法得到不断提高,高频数据的获得越来越便利。与低频数据相比,高频数据具有更为丰富的市场信息,因此成为目前的研究热点之一。随着进一步的研究和关注,量价关系在金融界的价值得到了较大的提高。股市的量价变化间既有着静态关系,同时也存在着较为明显的动态关系。本文的研究目的就是利用我国股票市场的历年高频数据,对股市量价动态关系进行深入探讨。首先在以往研究的基础上介绍了金融市场中的高频数据的基本概念和特征,并引入“已实现”波动;其次,对量价关系的四大理论模型进行了详细的阐述,这对后面采用高频数据进行实证分析有重要意义。虽然金融领域对于成交量与波动性间的关系研究一直是一个焦点,但以往学者对其研究主要关注于低频数据,所以利用高频数据实证研究交易量和波动性间的关系有一定的意义。在此基础上构建GARCH模型,结合ARMA结构的均值方程,构建了描述量价之间关系的ARMA-GARCH预测模型。基于DCC-GARCH模型进行参数估计和检验,同时利用高频数据对股市量价关系进行实证分析。结果表明,股价和成交量间的动态相关关系并不是常数,而且具有持续性和时变性,伴随着市场信息流呈现较强的波动性。
[Abstract]:In recent years, with the continuous development of computing technology and electronic trading system, as well as the decline of transaction costs, data acquisition and processing methods in the financial market have been continuously improved, and the acquisition of high-frequency data is becoming more and more convenient. Compared with the low frequency data, the high frequency data has more abundant market information, so it has become one of the research hotspots. With further research and attention, the value of volume-price relationship in the financial world has been greatly improved. There is both static and dynamic relationship between the volume and price changes of stock market. The purpose of this paper is to make use of the high frequency data of stock market over the years to probe into the dynamic relationship between volume and price of stock market. Firstly, this paper introduces the basic concepts and characteristics of high-frequency data in financial markets, and introduces the "realized" fluctuations. Secondly, the four theoretical models of the relationship between volume and price are described in detail. This is of great significance to the empirical analysis using high-frequency data. Although the relationship between trading volume and volatility has always been a focus in the financial field, previous scholars have focused on low-frequency data. Therefore, using high-frequency data to empirically study the relationship between trading volume and volatility has certain significance. On this basis, the GARCH model is constructed, and the ARMA-GARCH prediction model describing the relationship between quantity and price is constructed by combining the mean equation of the ARMA structure. The parameters are estimated and tested based on DCC-GARCH model, and the relationship between volume and price of stock market is analyzed by using high frequency data. The results show that the dynamic correlation between stock price and trading volume is not constant, and it is persistent and time-varying, which is accompanied by strong volatility of market information flow.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F830.91;F224

【参考文献】