| The vigorous development of China’s securities investment market has promoted the prosperity of securities research industry and the emergence of related professions.Securities analysts are professional analysts in the securities market.They provide market participants with price information that reasonably reflects the intrinsic value of securities through their superior information collection methods and professional analysis ability.Among the research reports published by analysts,investors are most concerned about the part of earnings forecast,and take it as an important basis for whether stocks have investment value.As analysts continue to closely track listed companies,their access to information will increase,and they will revise and update their earnings forecasts.Focusing on the gradually growing analyst group,this paper explores whether the analyst’s earnings forecast revision can be used as a pricing factor to enhance the ability of mainstream asset pricing model to explain excess returns,and tests its stock selection ability in actual trading by constructing quantitative investment strategies.This paper takes A-share market as the research object and selects monthly data from January 2008 to December 2021 for empirical test.Firstly,the empirical research on analyst earnings forecast revision and stock excess return is carried out,including univariate and bivariate portfolio analysis,Fama-Macbeth regression.Based on the ratio of market capitalization to book value,a 5×5 portfolio of sample stocks is constructed,and the Revise Minus Unrevise(RMU)is incorporated into the Mainstream multifactor modes like Capital Asset Pricing Model(1964),Fama-French Three-factor Model(1993),and Hou-Xue-Zhang Four-factor Model(2015)to explore its pricing ability by GRS Test,Fama-French Intercept Term Test,and Factor spanning Test.In addition,the trading is simulated by designing quantitative trading strategies,the results of Hierarchical Back Test are displayed and analyzed,and the comparative analysis are made by adjusting the calculation method of indicators.The empirical results show that:(1)The univariate portfolio test shows that there is a positive correlation between the portfolio return rate and the analyst earnings forecast revision,that is,the portfolio with the larger the analyst earnings forecast revision range in the current month can achieve higher excess return in the next month.In addition,after controlling the risk factors,the performance of the arbitrage portfolio is still stable.That is,there is a significant difference between the groups and there is a risk-free arbitrage opportunity,which confirms the investment value of analyst earnings forecast revision and verifies the existence of analyst earnings forecast revision anomaly.(2)According to the analysis of bivariate portfolio,after controlling the equity and financial characteristics of listed companies,such as market capitalization/book value ratio/ROE,the analyst earnings forecast revision premium still exists,which indicates that the analyst revision information contains other information except the company’s market capitalization/book value ratio/ROE.Moreover,after the adjustment of risk factors,the long-short hedging strategy based on the analyst’s profit forecast correction can still obtain significant excess returns.(3)Fama-MacBeth regression results show that in China Ashare market,analysts’ earnings forecast revision has positive predictive ability for stock crosssectional returns.By adding standard unexpected earnings(SUE)as a control variable,the research results show that analysts’ earnings revision’s predictive ability for future returns mainly comes from their judgment on the company’s future fundamental information.In addition,analysts’ earnings forecast revision does not have the ability to continuously forecast stock excess returns in the medium and long term,and the significant positive correlation between them only lasts for two months.(4)After adding the Revise Minus Unrevise(RMUs),the new model has stronger pricing power and independent pricing power than the mainstream multi-factor model.(5)Robustness test shows that the premium of analysts’ earnings forecast revision still exists after taking advantage of the information contained in the Schmidt orthogonal exclusion of standard unexpected earnings(SUE).In order to explore another part of the source of premium,this paper introduce the proportion of small capital net inflow(retail trading behavior)and the herd behavior of institutional investors as intermediary variables for further empirical analysis.The research shows that the correction of this information conveys a positive signal to institutional investors,and the behavior of group buying increases,while the behavior of group selling is reduced.In addition to institutional investors,the trading behavior of retail investors will also be affected by the analyst’s earnings revision,and the analyst’s earnings forecast revision will promote the proportion of small capital net inflow(the participation degree of retail investors)in the next period(t+1 period).The quantitative results show that this factor has a significant ability to obtain positive returns in the back test period,and the net worth curve of the first group to the fifth group is clearly divided,among which the cumulative return of the fifth group is 357.17%.The annual returns of each group are 5.01%,10.47%,13.59%,14.34%and 16.41%,respectively.In addition,the win rate of the long-short portfolio is 54.17%,the cumulative return rate is 167.79%(between the first group and the second group),and the annual return rate is 10.35%,and RMU stock selection has stability. |
