ABSTRACT

Recent SEC guidelines enabled many Fortune 500 companies to actively adopt social media, such as Twitter, to disseminate information. In this paper, we analyze the relationship between tweets by corporations and stock returns. Our study used over 1.2 million corporate tweets made by thirty companies in the Dow Jones Industrial Average between April 2013 and July 2020. The shocks from the frequency of corporate tweets can positively impact stock returns and trading volume. We, therefore, examine causality and impulse response between frequency of corporate tweets, stock returns, and changes in trading volume using a vector autoregression model. Our findings indicate that 43 percent of stocks exhibit Granger causality between firm-initiated tweets and changes in trading volume. We find evidence consistent with the attention-induced price pressure hypothesis proposed by Barber and Odean. We observe that a shock in corporate tweeting behavior translates into a positive effect on changes in trading volume and returns in 73 percent and 60 percent of stocks, respectively. These results are significant for developing appropriate social media communication strategies. The findings are also valuable for investors and traders who can deploy forecasting models utilizing corporate tweets to earn superior returns.

摘要

美国证券交易委员会最近的指导方针使许多财富 500 强公司能够积极采用 Twitter 等社交媒体来传播信息。在本文中，我们分析了公司推文与股票收益之间的关系。我们的研究使用了 2013 年 4 月至 2020 年 7 月道琼斯工业平均指数中 30 家公司发布的超过 120 万条公司推文。公司推文频率带来的冲击可以对股票回报和交易量产生积极影响。因此，我们使用向量自回归模型检查公司推文频率、股票收益和交易量变化之间的因果关系和脉冲响应。我们的研究结果表明，43% 的股票在公司发起的推文和交易量变化之间表现出格兰杰因果关系。我们发现证据与 Barber 和 Odean 提出的注意力诱导的价格压力假说一致。我们观察到，企业推文行为的冲击转化为对交易量和回报率变化的积极影响，分别为 73% 和 60% 的股票。这些结果对于制定适当的社交媒体传播策略具有重要意义。这些发现对于可以利用公司推文部署预测模型以获得丰厚回报的投资者和交易员也很有价值。