An Empirical Research On Asset Allocation Based On The Optimized Black-Litterman Model

Asset allocation has an extraordinary position in modern asset management.Carrying out reasonable and effective post-investment management is the issue most concerned by each market participant.Therefore,the optimal combination weights are obtained by constructing a mathematical model for asset allocation.Currently,the mainstream asset allocation strategies mainly include the Black-Litterman model and the style rotation model.How to reasonably and effectively use these models to construct configuration strategies for portfolio investment is extremely important for investors to reduce risks and obtain excess returns.Based on these two models,this paper empirically studies their application in the asset allocation of large categories and industries.This article begins with a systematic review of the theory of asset allocation and the latest research results at home and abroad.The third chapter is the theoretical modeling part of Black-Litterman model,and based on this part,theoretical improvement based on expert opinion and ARMA model was performed.Then the ARMA model in the time series method is used to optimize and improve the subjective view of the traditional Black-Litterman model,and the investor's subjective expected return rate is replaced by the predicted value of the asset's historical rate of return.And based on the improved Black-Litterman model,the stocks,bonds,commodity futures and other major assets are studied.The fifth chapter builds a four-quadrant wheel model based on relative strength index to analyze the industry assets.Selecting ten industries of Shanghai Securities as industry sample targets.In combination with market consensus,ROE is input as a subjective rate of return into the Black-Litterman model,and the results of the configurations with different confidence levels are compared.The empirical results show that,instead of the subjective view,the predictive value of the ARMA model can well avoid the disadvantage that the traditional view is arbitrary and can not be quantified,and the fitting value is better as the configuration result of the opinion rate of return.The dual-model industry configuration strategy based on style rotation and Black-Litterman can obtain consistently better optimal configuration weights,and the performance of the optimal combination of historical performance is better than that of the benchmark index,and can obtain a very considerable excess return.At the same time,adjusting the ratio of opinion errors can effectively improve the performance of the model.As a whole,the application of the Black-Litterman model in large-scale and industrial asset allocation is very effective.