不确定和随机环境下绿色投资组合问题的稀疏高次模型求解

Sparse higher moment model for green portfolio problem in an uncertain random environment

本文在不确定随机的混合市场环境中,针对绿色投资组合优化问题,建立了具有高阶矩的多目标稀疏模型,提出了一种绿色和非绿色资产混合配置的投资策略.具体地,本文把最近上市且缺乏足够样本信息的绿色资产作为不确定变量,而其他具有足够历史数据的资产作为随机变量进行统一建模.在这样一个混合资产环境下,建立了一个具有基数约束的均值-平均绝对下半偏差-偏度的投资组合优化模型,并采用快速非支配排序遗传算法对模型进行求解.最后,针对提出的模型与算法,对中国沪深股市的实际数据集进行了样本外分析.实证结果表明了本文提出的混合配置策略在Sharpe Ratio方面的优越性.

This study investigates a hybrid allocation benefits of green and non-green assets in an uncertain and random market environment by proposing a multi-objective sparse model incorporating higher-order moments.Specifically,we treat recently listed green assets with insufficient sample information as uncertain variables,while other assets with adequate historical data are modeled as random variables in a uniform manner.To address this uncertain and random environment,we propose a mean-downside-variance-skewness portfolio optimization model with a cardinality constraint.The fast nondominated sorting genetic algorithm(NSGA-Ⅱ)is employed to solve the model efficiently.Furthermore,we conduct out-of-sample empirical analysis using actual datasets from the China Shanghai-Shenzhen stock market to demonstrate the advantages of our hybrid allocation strategy.