Purpose

The authors study the explanatory power of investor rationality and irrationality for value and momentum portfolios. We also examine the relationships during financial crisis events, namely, the US subprime mortgage crisis (2007–2009) and the European debt crisis (2011–2013).

Design/methodology/approach

This study examines the influence of investors’ rationality and irrationality on the US stock market, using the multiple linear regression model and the stepwise regression model. Technically, the stepwise regression uses the machine-learning technique, with specific testing methods — forward selection, backward selection and stepwise selection — to find the best-fit model, according to Akaike’s Information Criterion (AIC). Thus, in this study, we will show the best model, as tested by the stepwise regression model.

Findings

Our empirical results contribute to the importance of reasons and emotions for stock-market returns and conclude that rationality and irrationality simultaneously explain the value and momentum portfolios, as well as the ETF portfolios. Also, the rational and irrational explanatory powers differ, depending on portfolios and different periods. Rational factors usually explain the volatility of the return to a greater extent than irrational factors. Moreover, during a financial crisis, the irrational factors remarkably increase their importance in explaining returns, especially for the ETF portfolios.

Originality/value

We expect this study’s contribution will show not only academic contribution but also benefit many stakeholders in the financial market. Investors and traders can identify various irrational factors of trading — for example, taking a long position during the panic in the market following the indicators in the models. Managers also reconsider the cost of the company by adding irrational factors when computing the equity’s expected return. Similarly, stock exchanges can adequately adjust their circuit breaker during a pessimistic-investor period. Finally, regulators can evaluate a complete picture of the stock market by adding irrational factors into their considerations.

中文翻译：

---

股票回报驱动因素：原因和情绪的混合

目的

作者研究了投资者理性和非理性对价值和动量投资组合的解释力。我们还研究了金融危机事件期间的关系，即美国次贷危机（2007-2009 年）和欧洲债务危机（2011-2013 年）。

设计/方法/途径

本研究采用多元线性回归模型和逐步回归模型考察了投资者的理性和非理性对美国股市的影响。从技术上讲，逐步回归使用机器学习技术和特定的测试方法——前向选择、后向选择和逐步选择——根据 Akaike 的信息准则 (AIC) 找到最合适的模型。因此，在本研究中，我们将展示最佳模型，并通过逐步回归模型进行测试。

发现

我们的实证结果有助于说明原因和情绪对股市回报的重要性，并得出结论，理性和非理性同时解释了价值和动量投资组合，以及 ETF 投资组合。此外，理性和非理性的解释力也不同，这取决于投资组合和不同时期。理性因素通常比非理性因素在更大程度上解释回报的波动。此外，在金融危机期间，非理性因素显着增加了它们在解释回报方面的重要性，尤其是对于 ETF 投资组合。

原创性/价值

我们期待这项研究的贡献不仅体现在学术贡献上，也将惠及金融市场的众多利益相关者。投资者和交易者可以识别交易的各种非理性因素——例如，在市场恐慌期间按照模型中的指标建立多头头寸。经理们还通过在计算股权的预期回报时加入非理性因素来重新考虑公司的成本。同样，证券交易所可以在悲观的投资者时期适当调整熔断机制。最后，监管机构可以通过在他们的考虑中加入非理性因素来评估股市的全貌。