Economic Cycle,Industry Rotation And A-share Market Investment Strategy

The industry's rotation in the stock market has long attracted the attention of the academic and investment communities.In particular,after the rise of quantitative investment,industry rotation has become a hot research topic.Intuitively,changes in industry profitability drive industry rotation.However,changes in industry profits are often related to fluctuations in the overall macro economy.Therefore,researchers linked the economic cycle to industry rotation in the stock market.By identifying the differences in the performance of industry stocks at various stages of the economic cycle,look for opportunities to profit from industry rotation.This paper systematically and comprehensively studies the interrelationship between macroeconomics,industry and stock market,tests whether the above intuition conclusions are established,and formulates feasible investment strategies based on the conclusions obtained.Finally,a backtest is performed to verify the correctness of the conclusions in this paper.This article is divided into three major parts.The first to third chapters constitute the first part,systematically studying the relationship between economic cycle,physical industry and industry rotation.This part of the research is mainly focused on the macroeconomic field.The fourth chapter constitutes the second part.Based on the integration of the important research conclusions of the first part,the concept of momentum and reversal effect is taken as the node,the economic cycle and the industry index are connected,and the real reason of the industry rotation is analyzed..This section transforms the research field from macroeconomics to the securities market.The third part of this paper consists of the fifth and sixth chapters.This part analyzes the possibility of profit from the industry's rotation strategy through design,implementation and evaluation.There are many literatures on the study of the economic cycle.Although the viewpoints and methods are different,the mainstream scholars have more consistent views on the stage of the economic cycle.The mainstream division of the current economic cycle stage mainly consists of the Dichotomy and the Quadrature.Compared with the quartering method,although the dichotomy is slightly rough,the identification method of the peak?valley?between the expansion?contraction?and the deflation?expansion?phase is simple and accurate.This paper first uses the dichotomy to identify the economic cycle turning point?ie,the peak or trough?,and on this basis,draw on the Merrill Lynch investment clock idea and divide the economic cycle by the Quadrature.When the economic cycle is divided,the Merrill Lynch investment clock only considers two factors,inflation and output gap,which will cause endogenous contradictions.In order to overcome this problem,considering the inflation and output gaps,consider the dichotomy results and output gap symbols,and introduce the concept of macroeconomic shock to divide the various stages of the economic cycle quartile.Based on the division of the economic cycle,Chapter 2 uses the consistency index to measure the synchronism between GDP and industry as well as industry and industry.A total of 75 industries participated in the calculation,so a totalC²₇₆?28?2850 consistency index?including GDP?can be obtained.According to the judgment of whether the consistency index is significant or not,a synchronous network between macro economy and industry,industry and industry is constructed.It can be seen from the network results that in the economic system,the synchronous relationship between the industry and the economic cycle is inconsistent,polarized and hierarchical,and there is no industry that has no indirect synchronization with GDP.The result indicates that the economic system acts as A whole,the link between industry and economy is inevitable and multi-layered.Also on the basis of economic cycle division,the third chapter studies the relationship between industry factors and industry rotation,economic cycle and industry rotation.First,the cointegration relationship between the industry's total turnover rate and the industry index was examined.Secondly,the SVAR model including the industry's total turnover rate,industry index,market index and interest rate is constructed.Finally,the industry index returns are statistically analyzed in different stages of the economic cycle.The results show that among the 75 industries,only 21 industries have a cointegration relationship between the industry's total turnover rate and the industry index,and only a few industry turnover rates have a large impact on the short-term fluctuations of the industry index.According to the performance of different industries in different economic cycle stages,the industry can be divided into cyclical industries and non-cyclical industries.As a transitional chapter,the fourth chapter,on the basis of combing the previous results,shifts the research perspective from the economic system to the financial market.When the stock market is weakly effective,the stock price reflects all historical information.Although the existing literature has no way to achieve weaker effectiveness in China's stock market,many documents confirm that the efficiency of China's stock market is constantly improving.This explains why only a few industry fluctuations affect the industry index.This result means that the securities market is melted by the investor's competitive arbitrage behavior due to the stable relationship between the industry and the economic cycle in history.If you want to profit from the industry's rotation phenomenon,the information provided by the changes in the industry's operating conditions is limited,and you can only find another way.Through the analysis of the momentum phenomenon of the industry in the A-share market.In essence,this paper argues that industry rotation is an industry momentum or reversal effect based on the economic cycle stage.Therefore,grasping the profit opportunities of industry rotation should start from two aspects:First,timely and accurate identification of the stage of the economic cycle at the starting point of investment;Second,based on the momentum and reversal effects of the industry under different economic cycle stages Characteristics,develop appropriate momentum investment strategies.Based on the above considerations,the fifth chapter studies the real-time identification method of economic cycle turning points.This paper believes that the real-time identification of the economic cycle turning point is essentially a kind of predictive behavior,so the research ideas and solutions are different from the first chapter.The change in the economic aggregate indicator is the final reflection of the mutual influence and interaction of many macroeconomic factors.Real-time identification of turn points using a single aggregate indicator clearly loses accuracy due to the loss of too much information.Therefore,the real-time recognition process of the economic cycle turning point is divided into two parts:the turning point identification of historical data and the real-time identification of turning points.In the historical data identification section,consistent with the first chapter,the BB model is still used to identify the dichotomy of the single economic aggregate indicator.In the real-time identification part,other important macroeconomic variables are introduced,and the LVQ algorithm is trained using the turning point recognition result of the historical data,and the trained algorithm is used for real-time recognition.The repeated test comparison shows that the real-time recognition scheme is superior to the traditional recognition method in recognition accuracy,recognition timeliness and recognition robustness.The sixth chapter designs an industry-based investment strategy based on the previous research conclusions and conducts a backtest.The strategy screens industry and stocks to build a basic portfolio based on the characteristics of industry momentum and reversal effects at each stage of the economic cycle.Taking into account investment risks,the Black-Litterman model is used to optimize the underlying portfolio.The investor opinion matrix in the Black-litterman model is determined by the GJR-GARCH-M model.The confidence vector is based on the classification indicator in the real-time identification of the economic cycle.In the backtesting process,a known and unknown data environment at the time of investment is strictly defined.Based on the latest macroeconomic variables released each month,the real-time results of the economic cycle turn point identification are continuously updated.Based on this,the industry momentum and reversal effect characteristics of the economic cycle are tested,and then the stock of the portfolio is adjusted according to this feature.Finally,the final combination weight is determined according to the BL model.The results of the test show that in most months,the strategy's return is better than the gain of the current month's market index.The research in this paper shows that the economic cycle has an impact on the industry's rotation in the stock market,but the main intermediate variable is not the industry factor.While cyclical fluctuations in the economy can trigger simultaneous or non-synchronous fluctuations in the industry,industry volatility has limited impact on industry indices in the stock market.Industry rotation is more of an industry performance of momentum and reversal effects.The link between the economic cycle and the industry index in the stock market is manifested by the changes in the characteristics of the industry's momentum and reversal effects at different stages of the economic cycle.In the process of research on the relationship between economic cycle and industry rotation,this paper makes some innovative attempts on the division of economic cycle,the investigation of the relationship between macroeconomics and industry,and the real-time identification of economic cycle.Due to the limitations of the understanding of the concept of industry rotation,this paper fails to make more specific and quantifiable improvements to this definition.The traditional definition that can only be used,which makes this article only limit the observation of the overall data when examining the relationship between industry rotation,economic cycle and industry,so that it can not go deep into the micro level of the specific“one”industry rotation phenomenon..This problem is not only the inadequacy of this paper,but also an important direction for future research.