Research On Asset Allocation Of Black-Litterman Model Based On Merrill Lynch Investment Clock

Asset allocation is an important branch of modern portfolio theory,national and regional configuration,scale configuration,style configuration and industry configuration constitute four aspects of tactical asset allocation adjustment,appropriate asset allocation can not only increase investment income but also avoid risks.Study found that the performance contribution of asset allocation to portfolio is more than 90%,among which the contribution of industry allocation is more than 20%,industry allocation strategy has become the mainstream analysis method.But at present the analysis of industry configuration is mainly qualitative analysis,with the development of the fund market,quantitative research of industry configuration get more and more attention,so this paper combination qualitative analysis and quantitative analysis method to study industry allocation problem.Merrill Lynch Investment clock model Study the asset allocation problem based on division of the economic cycle,which provides a new way for the industry configuration,this paper applicate "investment clock" model in China A stock market,to explore China’s sector rotation.The mean-variance model of Markowitz is the most classical portfolio model,and the Black-Litterman model proposed by Goldman Sachs in 90s is an improvement on the Markowitz model,It introduces the subjective view of investors and the confidence level of the subjective view into the model,which makes the Markowitz model more practical.In this paper,we compare the allocation efficiency of Markowitz model and Black-Litterman model with contrastive analysis method to obtain the optimal asset allocation ratio.The Black-Litterman model takes investor’s subjective view into account.The subjective view of the investor is the important parameter of the Black-Litterman model,The innovation of this paper is to use the analysis result of investment clock as the source of subjective view of Black-Litterman model,to quantify investors’subjective view and improve the accuracy of the model.In addition,this paper adopts the shrinkage method to calculate the variance covariance matrix,which overcomes the problem of the variance covariance matrix.In this paper,we take the excess return of GICS I level industry from June 2014 to June 2016 as the object of study,forecast the asset allocation in July 2016.According to the theory of business cycle,the period of economic cycle in July 2016 is judged as the overheating period.Based on the conclusion of the "investment clock" model,the relative performance of various industries in the overheated period is obtained,take it as the subjective view of Black-Litterman model to obtained the expected return of Black-Litterman model.Put the expected return rate of Black-Litterman model into the Markowitz model to obtain the optimal asset allocation ratio,compare optimal asset allocation of Black-Litterman model and Markowitz model,so as to better illustrate the effect to the proportion and efficiency of the asset allocation after the addition of the investors’ subjective perspective.Empirical analysis results show that the Black-Litterman industry allocation retuen of 10%,50%,80%confidence level are all greater than the market portfolio and the Markowitz return.And the higher the level of confidence,the greater the return of Black-Litterman model industry configuration,which is sufficient to show the superiority of the Black-Litterman model.In addition,although the SHARP ratio of Black-Litterman models with three different confidence levels are negative,but still higher than the SHARP ratio of market portfolio and Markowitz portfolio,shows that when the market fell,the unit risk of Black-Litterman asset allocation brings smaller losses.This paper discusses the industry configuration based on the investment clock,but due to the length of the space,the allocation strategy of the industry from other way is not studied,so the framework of industry configuration strategy is not constructed perfectly.Due to the large amount of information involved in the operation of the industry configuration and the highly correlated date,How to choose an effective processing mode,especially how to deal with the difference between the basic information and market expectations can be thinking deeply.In addition,the domestic market structure varies greatly,if the outcome deponding on the historical test can be directly used in the future market remains to be further tracking and exploration.