Research On The Correlation Between Macroeconomic Factors And Stock Market ?s Based On MIDAS Framework

The CAPM model proposed by William Sharp decomposes the risks of investing a stock into systemic risks and non-systematic risks.The systematic risk can be abbreviated as ?,which is of homogeneity in the stocks with some similar characteristics.Preventing systemic financial risks is an eternal theme of the capital market,indicating that studying the source of systemic risk of the stock market is of strategic significance for maintaining national security,improving the financial system's ability to serve economic activities,and shrinking financial bubbles.We all know that macroeconomic fluctuation is one of the major sources of systemic risk.In this way,this article defines macro factors as the key indicators that affect the stock price except for the microstructure of the capital market,and which are transmitted through systemic risk.They can reflect the overall economic condition of a country at an aggregate level.In order to prevent systemic risks and improve the pricing system,we think that it is necessary to explore the transmission and magnitude of specific macro factors on stocks' ?.Previous studies have proved that macroeconomic conditions such as inflation,economic growth,monetary policy,market expectations,and risk aversion can increase or decrease the stocks' ? by affecting the discounted cash flow of enterprises and people's asset allocation strategy.But so far,there is little theoretical research on systemic risks,and empirical research is widely used in this field.The mixed data sampling(MIDAS)proposed by Ghysels in 2003 can increase the data capacity of samples and incorporate more high-frequency data into the regression models and shows better out-of-sample test ability.It is till now widely used in the research fields such as macro financial forecasting and stock price fluctuation.Therefore,this paper combines the conditional capital asset pricing model with the MIDAS framework,and uses the consumer price index,the producer price index,the industrial added value,money supply of narrow sense(M1),the one-year and five-year government bond maturity yields,the five-year corporate bond maturity yield to construct macro factors.We attempt to decompose the stock ? into a short-term and a long-term component.We use the high-frequency stock market historical returns to calculate the short-term component of ?s,and define the long-term component as a linear function of macro factors.As for constructing the investment portfolios,we refer to the Fama-French three-factor and five-factor models.After excluding all finance companies in Chinese stocks,we separate the other stocks into two groups based on two dimensions:total market value and book-to-market ratio.The empirical results show that the term spread has a significant effect on the? of the big-size portfolio.Also,the model has a good performance in R2 and other coefficients' significance.When the term spread increases,the big-size portfolio's mixed frequency ? seems to decrease.We analyze the four situations that can affect term spreads,and found that the big-size portfolio's ? is counter-cyclical,which is consistent with several previous findings.The high-frequency ? of big-size portfolio has a smaller weight in the mixed-frequency ?,but it explains more than 70%of the fluctuation.The maturity of long-term bonds has little effect on the empirical results in our robustness test,ensuring that the term spread a good variable for explaining the time-varying systemic risk of big-size stocks.Term spread is affected by multiple factors such as inflation,economic growth expectations,people's risk aversion,monetary policy and so on.However,this paper's empirical results show that CPI,PPI,money supply,the industrial added value,the interest rate of one-year government bonds don't have a significant influence on the mixed-frequency ?s.It seems that single indicators regarding to monetary policies and market liquidity can hardly explain the time-varying ?s during business cycles.we think subsequent studies can explore the correlation between term spreads and other macro factors used in this paper to characterize the time-varying ? more clearly.Supervisors are welcome to refer to this paper when making or implementing policies.For investors,this paper can guide to understand the time-varying characteristics of stocks' ?.Investors can also use our research to optimize the asset allocation and strengthen risk management.