Black-Litterman模型的参数优化及其在行业资产配置中的应用

Parameter optimization of Black-Litterman model and its application in industry asset allocation

针对Black-Litterman模型中投资者观点的量化问题,运用梯度提升回归树(gradient boosting regression tree,GBRT)算法对收益率进行预测,将预测值作为观点收益的替代变量,算法迭代收敛时的均方误差作为观点误差。采用上证380的10个行业指数数据检验文中提出的参数优化方法的有效性及合理性。结果表明,结合GBRT算法的Black-Litterman模型的投资绩效优于市场策略和Markowitz模型。

For the problem of the quantification of investor views in the Black-Litterman model, this paper proposes to use gradient boosting regression tree (GBRT) algorithm to predict the returns, regarding the predictive values as view returns and take the mean square errors when the algorithm converges iteratively as the view errors. The data of Shanghai 380 industry index are used to test the validity and rationality of the parameter optimization method proposed in this paper. Results show that the investment performance of Black-Litterman model combined with GBRT algorithm outperforms the market strategy and Markowitz model.