基于ACD-UHF-GARCH模型的中国股票市场波动性研究

发布时间：2017-12-31 01:13

本文关键词：基于ACD-UHF-GARCH模型的中国股票市场波动性研究　出处：《东北大学》2013年硕士论文　论文类型：学位论文

【摘要】：在市场经济的条件下,中国股票市场越发成熟。目前,股票市场的波动性研究一直是金融研究的热点。目前,很多学者对股市量价关系进行了研究,但是很少引入时间因素,因此本文将交易量和时间因素同时纳入研究的模型中,解释持续期和交易量在价格波动中所起的作用。应用高频数据研究股市波动性实质上是应用交易时间间隔来测定波动性。高频时间序列包含市场微观结构的信息及重要的长期日间现象的信息。本文在以往研究的基础上,采集沪深两市逐笔交易和等间隔交易的数据,选用研究高频数据的ACD模型和在应用低频数据测度股市波动性表现较好的GARCH模型,应用ACD模型计算出交易持续期均值,再应用所得参数构建处理超高频数据的GARCH模型,即UHF-GARCH模型。在建立UHF-GARCH模型时,本文不仅考虑到交易的持续期,同时还考虑到交易量对股市收益率和波动性的影响。本文分别选取了个股数据和指数数据进行实证研究。个股数据选取浦发银行和平安银行2012年11-12月逐笔交易的超高频数据,指数数据选取上证综指和深证成指2011年和2012年两年的每分钟高频数据。实证研究表明：中国股票市场的交易持续期和收益率的波动性具有聚类性的特征；交易持续期对股市波动率有影响,持续期越大,波动率越大；同时可以得出交易量对波动率的影响更显著,交易量越大,波动率越大。说明ACD-UHF-GARCH模型可以很好的反映高频收益波动聚类性的本质,持续期和交易量对价格波动有动态影响。
[Abstract]:Under the condition of market economy, the stock market in China is more mature. At present, the volatility of stock market has been a hot topic in financial research. At present, many scholars have studied the relationship between volume and price of stock market. However, the time factor is rarely introduced, so the transaction volume and the time factor are included in the model. Explain the role of duration and trading volume in price volatility. The use of high-frequency data to study stock market volatility is essentially the use of trading time intervals to measure volatility. The high-frequency time series contains information about the market's microstructure. And important information on long-term daytime phenomena. This paper is based on previous studies. To collect the data of Shanghai and Shenzhen stock markets, we choose the ACD model to study the high frequency data and the GARCH model to measure the volatility of the stock market by using the low frequency data. The ACD model is used to calculate the average transaction duration, and then the GARCH model to deal with UHF data is constructed by using the parameters. That is, UHF-GARCH model. When establishing UHF-GARCH model, this paper not only takes into account the duration of the transaction. At the same time, considering the influence of trading volume on the stock market return and volatility. This paper selects the data of individual stocks and index data to carry on the empirical research. The data of individual stocks are selected from Pudong Development Bank and Ping an Bank on 2012 11-1. UHF data traded on a case-by-case basis in February. The index data are selected from the Shanghai Composite Index and Shenzhen Composite Index on 2011 and 2012. The empirical study shows that:. The volatility of trading duration and yield in Chinese stock market is characterized by clustering. Trading duration has an impact on the volatility of the stock market, the longer the duration, the greater the volatility; At the same time, it can be concluded that the effect of trading volume on volatility is more significant, and the greater the volume, the greater volatility. It shows that ACD-UHF-GARCH model can well reflect the nature of high frequency return volatility clustering. Duration and trading volume have a dynamic impact on price volatility.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.51;F224

【参考文献】