| The capital asset pricing model is the main content of modern financial economics,The aim is to study and determine the price of assets that are uncertain future earnings from a risk-benefit perspective.The 21st century will be the era of large data,the data will be the most powerful tool,and thus the investment thought or ideas transformed into mathematical models,with large data on the real situation for back analysis,will become a reasonable and effective way.The latest research results of the asset pricing model are:Based on the dividend discount model in 2015,the Fama-French five-factor model suggests that the profit factor and the investment factor are added on the basis of the original three-factor model,which improves the economic theory of the model Foundation,goodness of fit and ability to interpret stock returns.China's stock market was established in 1990,is a typical emerging market,and China's stock market was born more than Europe and the United States stock market for more than a century,and thus apply to the European and American markets for asset pricing model is applicable to the Chinese stock market,whether for our country The stock market service is a meaningful study,which is very important for the investment decisions of government regulators,enterprises and individual investors.At the same time,it can also use the asset pricing theory to understand and explain a certain risk The historical price or the level of the asset.Based on the theory and method of domestic and foreign mature asset pricing research,this paper studies the effectiveness of Fama-French five-factor model in China stock market from two dimensions of cross-sectional and time series.This paper selects between January 2002 to December 2016,All A share stocks on the stock market of Shanghai and Shenzhen as a research sample,eliminate the new listing for 3 months,6 months before and after the IPO,ST,PT,financial stocks in sequence as qualified stocks.Firstly,the analysis ofcross-section regression of the Fama-French two-step method was used to study the interpretation ability of the difference in cross-sectional yield of individual stock insingle,combined or joint after 1260 cross sectional regression.Then,the time series regression analysis is carried out to construct the portfolio simulation model.The explanatory variables and the influencing factors are sorted and grouped respectively.The annual samples are sorted by market value,divided into five groups from small to large,The market value ratio,profit factor and total asset growth rate are further subdivided into five groups,25 market value-book value ratio combination,25 market value-profit factor combination,25 market value-investment factor combination,and descriptive statisticsof the excess return yield of 75 stock portfolios.At the same time,the market factors,market value factor SIZE,value factor HML,profit factor RMW and investment factor CMA were constructed by 2*3 bivariate sorting at the end of each month,and descriptive statistics and correlation analysis were carried out for 5 factors.Finally,25 market value-book market value ratio combination,25 market value-profit factor combination,25 market value-investment factor combination of time series regression,respectively,intercept and 5 factors regression coefficient,t statistics,and finally through the GRS Test,the statistical intercept is not zero to verify the zero factor of the five factor model,and its stock portfolio yield time series differences in the interpretation of the ability.The results of this paper show that the value factor and the market value factor have high significance,and have a great influence on the cross-sectional profit of the individual stocks,and the profit factor and the investment factor have little effect on the cross-sectional profit of the individual stocks.The Fama-French five-factor model can explain the time series change of the A-share market returns in Shanghai and Shenzhen,and the scale factor has a strong ability to explain the change of the stock portfolio and is negatively correlated with the stock portfolio yield.Factor regression coefficient is stable at 1;the value factor HML is more significant in large scale,high profitability,low asset growth group,and the regression coefficient is positive and negative,can not determine the direction of the impact of the yield;the profit factor RMW has a positive effect on the return of stock portfolio.The investment factor CMA is significantly higher in the combination of low profitability and total assets growth,and most of the regression coefficients are negative,indicating that the investment factor has a positive effect on the yield of the stock portfolio,Which is the opposite of the US stock market.Although the regression factor of the influencing factor is not all zero,the five-factor model is generally applicable in China's A-share market,and it also encourages us to excavate new vision variables in the light of the actual situation in China and find a more perfect Asset pricing model.The main innovations of this paper are two:First,the use of the latest quarterly data and mid-year report data to build pricing factors,net profit attributable to shareholders of the parent company using single-quarter income statement data;total assets growth rate of the first quarter,mid-year,third quarter,annual report data,And the equity attributable to the shareholders of the parent company adopts the data of the mid-year report and the annual report.Each variable adopts the latest data to correctly match the stock's monthly yield(explanatory variable)with the financial indicator(explanatory variable);The monthly rate of return is always matched with the latest financial data that can be obtained in each month,which can be reflect the latest situation of listed companies,while making the research conclusions more rigorous.The second is to sort all stocks at the end of each month.As the quantity and performance of the stock in each month is inconsistent,a stock may be delisting in a certain period of time,ST,PT or data missing,and therefore for the accuracy of the conclusion,this month at the end of each month on the Shanghai and Shenzhen all A Stock stocks to filter,making the number of stocks per month inconsistent.qualified stock quantity is different in every month,and in this paper,using the latest quarterly data and data center daily news,so this paper innovation at the end of each month for all qualified stock sorting group,makes every stock portfolio consists of qualified stocks,and uses the latest data to construct the pricing factor,so that the research conclusion is more credible. |
PDF Full Text Request