Abstract
As a "barometer" reflecting the macroeconomic operation status,the stock market
plays a vital role in optimizing resource allocation and preventing and controlling
market risks. But compared with the mature and stable stock market, China's stock
market is still an emerging market,still accompanied by high uncertainty and
instability in the development process.Macroeconomic variables are considered to be
the important factors that cause the stock market fluctuation.Studying the influence of
macroeconomic related indicators on the stock market plays a very important role in
promoting the healthy and stable development of China's stock market.
This paper expands on previous research by using subsample average realized
volatility rAVGRV instead of realized volatility RV in addition to macroeconomic
variables when estimating the long-term volatility component of the GARCH-MIDAS
model. It takes into account not only the impact of macroeconomic variables on the
long-term volatility of the stock market, but also the impact of high-frequency stock
data information and market microstructure noise on the volatility of the stock market.
In the empirical analysis, the single and double factor GARCH-MIDAS models are
firstly established to investigate the impact of realized volatility and macroeconomic
variables on stock market volatility; Then, in order to study the impact of the overall
macroeconomic situation on the stock market fluctuation, the principal component
analysis method was further adopted, constructed two main components and one
comprehensive macroeconomic index, and systematically examined the impact of the
overall macroeconomic situation on the stock market fluctuation.Then,the
GARCH-MIDAS model constructed in this paper is compared with the prediction
effects of other models by means of loss function and MCS test to determine the pros
and cons of the model. Finally, the slightly expanded GARCH-MIDAS model is
applied to the investment portfolio, the economic application value of the model is
studied, and the robustness of the model is tested.
This paper selects the daily return data of the CSI 300 Index from April 8, 2005
to June 30, 2021, the realized volatility estimated based on 5 minutes high-frequency
data, and the sub-sample average realized volatility as the stock market data. The
monthly data of 10 representative macroeconomic variables, such as the consumer
price index, industrial added value, and total retail sales of consumer goods, are used
as macroeconomic sample data. The research results show that: the short-term
volatility of my country's stock market has a significant volatility aggregation effect;
IV

in the research on the influencing factors of the long-term volatility components of the
stock market, the influence of noise in the high-frequency stock data cannot be
ignored; and compared with the single-factor model, the two-factor model The results
in terms of fitting effect, predictive power and investment portfolio are relatively
good. It shows that the GARCH-MIDAS model with richer data information has
higher prediction accuracy and can enable investors to obtain higher economic value;
the selected macroeconomic variables have a significant impact on the long-term
volatility of the stock market. Among them, industrial added value and import and
export volume have the greatest impact on the stock market, and the exchange rate of
USD converted to RMB has a significant negative impact on the long-term fluctuation
of the stock market. In addition, this paper will combine the research results with the
specific reality, and put forward relevant policy suggestions from three aspects:
macro-control, financial supervision, and investor quality.
Keywords: Stock market volatility; GARCH-MIDAS model; principal component
analysis; MCS test; portfolio
V

目录
摘要.................................................................................................................................I
Abstract......................................................................................................................... II
1绪论............................................................................................................................1
1.1研究背景...........................................................................................................1
1.2研究意义..........................................................................................................2
1.2.1理论意义................................................................................................2
1.2.2现实意义................................................................................................2
1.3文献综述..........................................................................................................2
1.3.1同频数据下宏观经济影响股市波动的研究........................................2
1.3.2混频数据下宏观经济影响股市波动的研究........................................4
1.3.3文献述评.................................................................................................5
1.4研究内容和研究方法......................................................................................7
1.4.1研究内容................................................................................................7
1.4.2研究方法................................................................................................9
1.5研究的创新与不足........................................................................................10
1.5.1创新之处...............................................................................................10
1.5.2不足之处...............................................................................................10
2宏观经济影响股市波动的相关理论......................................................................11
2.1股票市场定价理论........................................................................................11
2.2宏观经济对股市波动的影响........................................................................11
2.2.1经济增长对股市波动的影响..............................................................11
2.2.2货币政策对股市波动的影响..............................................................12
2.2.3财政政策对股市波动的影响..............................................................13
2.2.4通货膨胀对股市波动的影响..............................................................13
2.2.5汇率与国际收支对股市波动的影响..................................................14
2.2.6消费与储蓄影对股市波动的影响......................................................15
2.3本章小节........................................................................................................16
3实证方法及模型简介..............................................................................................17
3.1混频数据抽样（MIDAS）模型...................................................................17
3.1.1 MIDAS模型的概述.............................................................................17
3.1.2 MIDAS模型权重函数.........................................................................18
3.2 GARCH-MIDAS模型 ...................................................................................18
3.2.1股市波动率成分分解理论..................................................................18
。。。以下略