基于损失厌恶和模糊厌恶的分布鲁棒投资组合模型

从行为金融学的角度,考虑投资者的损失厌恶和对资产收益均值的模糊不确定特征,研究风险资产收益分布未知条件下的分布鲁棒投资组合问题.假设收益率均值属于椭球不确定集,构建基于损失厌恶和模糊厌恶的鲁棒投资组合模型,并得到满足最坏可能分布的最优解.进一步应用数值算例分析投资者的动态损失厌恶特征及不同的初始损失厌恶和模糊厌恶程度对模型的影响.结果表明:在模糊中性和模糊厌恶条件下,积极投资者在最优期末财富方面的投资绩效优于保守投资者;损失厌恶和模糊厌恶系数越大,投资者的最优期末财富越低.

Research on distributional robust portfolio optimal model based on loss aversion and ambiguity aversion

From the aspect of behavioral finance, considering the loss aversion and fuzzy uncertainty of the asset mean return of investors, we research on distributional robust portfolio problem under the hypothesis that the distribution of asset return is unknown. We construct robust portfolio model based on loss aversion and ambiguity aversion character providing that mean return vector belongs to ellipsoidal uncertainty set. Then, the optimal solution is derived which meets the worst possible distribution. Furthermore, we empirically analyze the effect of dynamic loss aversion character and different degrees of initial loss aversion and ambiguity aversion on the model by use of numerical cases. The conclusions are that under both the ambiguity neutral and ambiguity averse conditions, active loss aversion investors outperform the conservative loss aversion investors from the aspect of optimal terminal wealth; the larger the loss aversion and ambiguity aversion coefficients are, the lower the optimal terminal wealth of investors would be.