羊群行为与股价波动的关系:来自理论层面的证据

Relationship between Herding Behavior and Stocks′ Price Volatility:Evidence from a Theoretical Perspective

随着行为金融学研究的蓬勃发展,股票市场中普遍存在羊群行为已经成为共识,但已有研究大多停留在实证层面,且对于羊群行为与股价波动之间关系的讨论并未形成一致的结论,部分研究认为羊群行为会加剧股价波动,还有研究表明羊群行为可能促进股票市场的稳定。从理论层面分析羊群行为和信息状态与股价波动之间的关系。基于无套利原理和期权博弈方法,构造资产组合复制正常交易中的股票,以波动性价值描述波动在股票价格构成中的贡献率,并将其作为测量股价波动的指标。基于信息视角构建羊群行为模型,描述羊群行为和信息状态对股价波动的作用,建立羊群行为视角下的股票波动性价值模型。运用最小二乘蒙特卡罗方法求解模型,具体分析不同信息条件下羊群行为对股票波动性价值的影响。研究结果表明,羊群行为对股价波动的影响需要根据不同的信息条件进行分析。羊群行为加剧市场信息向完全"知情"和完全"不知情"分化,因而在信息不完全但对称的条件下,低信息水平的市场中羊群行为对股价波动产生收敛作用,高信息水平的市场中羊群行为对股价波动产生发散作用,而在中等信息水平的市场中羊群行为的作用则不明显。对于信息不完全且不对称的市场,表现为在同等信息不对称条件下,羊群行为在信息优势不明显时平抑股价波动,而在信息优势足够大时加剧股价波动。研究结果对实证研究中羊群行为是否加剧股价波动的分歧观点给出合理的解释,进一步完善了现代资产定价理论;为监管部门健全信息披露制度、加强投资者教育引导等政策措施提供理论依据,帮助投资者充分认识市场信息状态和有限理性行为,为优化决策、规避风险提供参考。

With the vigorous development of behavioral finance research, the prevalence of herding behavior in the stock market has become a consensus, but unfortunately, most of the existing research remains at the empirical level, and the discussion of the relationship between herd behavior and stocks′ price volatility fails to reach a consistent conclusion. Some studies suggest that herding behavior will exacerbate the volatility of the stocks′ price, while the others show that herding behavior may accelerate the stability of the stock market.This study analyzes the relationship between herd behavior, information status and stocks′ price volatility from a theoretical view. Based on the principle of no-arbitrage theory and the idea of option game, an asset portfolio is constructed to replicate the stocks in the normal trading, the stocks′ volatility value model is established to describe the contribution of volatility in the composition of stocks′ price, and it is used as an index to measure stocks′ volatility. Basing on the information-based herd behavior model, the study describes the effect of herd behavior and information state on stock price fluctuation, and constructs the stocks′ volatility value model driven by herd behavior. Using the Least Square Monte Carlo method, the study analyzes the impact of herd behavior on stocks′ volatility value under different information conditions.The results of the study show that the impact of herding behavior on stocks′ volatility needs to be analyzed based on different information conditions. As herding behavior will exacerbate the differentiation of market information towards fully "informed" and "uninformed", under the condition of incomplete but symmetrical information, herding behavior in the market with low information level will have a convergence effect on stocks′ volatility. Herding behavior in the high-informed market will have a divergent effect on stocks′ price volatility, while the role of herding behavior in the medium-informed market is not obvious. For markets with incomplete and asymmetric information, it is shown that when the equivalent information advantage is not obvious, market volatility can be stabilized, and when the equivalent information advantage is large enough, the herding behavior will exacerbate stock market volatility.The results for this study provides reasonable explanations for the different opinions on whether herding behavior will exacerbate stock market volatility in current empirical researches, and further improves modern asset pricing theory. The relevant research results may provide a theoretical basis for regulatory authorities to improve information disclosure systems and strengthen investors′ education and guidance. It can also help investors fully understand market information status and bounded rational behavior, and provide reference for optimizing decision-making and avoiding risks.