Research On Black-Litterman Asset Allocation Model With GARCH Model And Viewpoint Weight

With the deepening of financial globalization in China,the innovation and upgrading of securities asset allocation model has become a hot topic in recent years.At present,the asset allocation model of China’s financial market mostly stays in the traditional asset allocation stage in the framework of mean variance.It is imperative to learn from and introduce the international asset allocation model,optimize and innovate.Under the framework of mean variance,the traditional asset allocation model mainly includes Markowitz portfolio model(1952)and Black-Litterman asset allocation model(1992),which has evolved from the model under strict assumptions.Compared with Markowitz portfolio theory model,the main innovation of Black-Litterman model is to "introduce the subjective view of investors",integrate the objective and subjective parameters,effectively make up for the shortcomings of the original model,and play a role of revision.However,in the application,Black-Litterman model does not fully explain the determination method of "part of model parameters",especially the content of "investors’ viewpoint quantification and parameter measurement" is less,so the practicability and accuracy of Black-Litterman model is weakened.On the basis of traditional asset allocation model,this paper firstly makes quantitative analysis on the subjective evaluation content of Black-Litterman model.On the one hand,GARCH model is introduced to estimate the view vectors and view error items of various assets,so as to obtain quantitative results in the asset allocation model;on the other hand,the estimated "market weight" of various risk assets is directly attached to the "n-dimensional line vector of investor’s view",so as to make the subjective view quantification more accurate and practical.So far,a new model has been formed: "GARCH-P Black-Litterman" securities asset allocation model.Secondly,expand the scope of empirical research to the comprehensive market at home and abroad,and expand the research field.Through the application of Eviews,Excel and other software,the use of OLS,ADF,arch and other test methods,the Sharpe ratio value is used to measure the configuration efficiency.The results show that: in different environments,the Sharpe ratio value of GARCH-P Black-Litterman model is higher than that of traditional asset allocation model,and the higher degree is different,about 0.01-0.1.The empirical results show that under the same conditions,the performance of GARCH-P Black-Litterman model is better.