Asset Allocation Based On Improved Hierarchical Risk Parity

With the rapid development of quantitative investment,investors gradually tend to hedge risks and increase returns by increasing the size of their portfolios,and asset allocation has become an important part of the investment process.The main issues studied in this paper are determining the optimal size of the portfolio and optimizing asset allocation.In this paper,various asset allocation models such as inverse fluctuation and mean-variance are introduced,and an asset allocation model based on the improved hierarchical risk parity algorithm is proposed on the basis of the existing hierarchical risk parity model,which increases the prediction steps of the correlation matrix between assets and improves the recursive dichotomous algorithm in the model.Firstly,the optimal number of securities in the portfolio is determined,the total risk of different sizes of portfolios is calculated by the precision method and the approximate method,and a reasonable asset quantity range is given.Secondly,based on the DCC-GARCH algorithm to predict the correlation matrix between assets,the hierarchical clustering algorithm is used to divide the assets into different groups,and diagonalize the assets to put similar assets in similar positions.Finally,the recursive dichotomous algorithm is used to assign weights to each asset in the portfolio,evaluate their performance,and bring in data analysis and comparison at different stages and different markets to obtain the optimal asset allocation model.The results show that in terms of risk,the risk level of the portfolio corresponding to the improved hierarchical risk parity model is slightly lower than that of the improved model,and lower than that of other basic models in this paper.In terms of returns,the improved model significantly improves portfolio returns as well as risk-adjusted returns.In summary,the improved risk parity model can significantly improve the robustness of the model,and effectively improve the portfolio income ability while reducing portfolio risk,which is also of certain reference significance for asset allocation in different capital markets in the future.