过度自信、后悔厌恶对收益率分布影响的数值模拟研究

通过构建数值模拟模型,对投资者受过度自信与后悔厌恶影响下的收益率分布进行模拟,结果证实:与有效市场假说之下的正态分布相比,此种状态下的收益率分布存在着尖峰厚尾、左偏、左尾厚于右尾以及尖峰与厚尾程度随着时间单位的延长呈现下降趋势等统计学特征.进一步的数值模拟还发现,随着反应滞后时间的增加,收益率分布峰度增加明显,左尾有下降的趋势;随着反应不足程度的增加,收益率分布的峰度与偏度同时增加;随着过度反应程度的增加,收益率分布右尾有着明显的增加;收益率分布右尾对处置效用不敏感,左尾随着处置效用的增加厚尾程度减少.

Numerical Simulation for Influence of Overconfidence and Regret Aversion on Return Distribution

According to behavioral finance,investors' behavioral biases will be reflected in the distribution of the return rate.This paper constructs a numerical simulation model to simulate the return rate distribution under the influences of overconfidence and regret aversion and shows that,compared with normal distribution under the Efficient Market Hypothesis,the simulated distributions have higher peaks and fatter tails,and they are skewed to left with the left tails thicker than the right ones.Moreover,the heights of peaks and the thickness of tails will decrease as the time horizons for the returns become longer.Further numerical experiments illustrate that,when the time lag of reaction increase,the kurtosis of the simulated distribution has an obvious increase,and the left tail tends to decrease;when under-reaction aggravates,the kurtosis and the thickness exhibit a concurrent increase;when the degree of over-reaction becomes large, the right tail has an obvious increase;the right tail is insensitive,while the thickness of the left tail tends to decrease with the increasing of the disposition effect.