The asymmetric online talk effect. (5th February 2021)

• Record Type: Journal Article
• Authors: Zhang, Ying; Lu, Xing; Prombutr, Wikrom
• Is Part Of: Review of behavioral finance. Volume 14:Number 2(2022)
• Page Start: 157
• Page End: 182
• Publication Date: 2021-02-05
• Subjects: Stock message board -- Online hype -- Neutral sentiment -- Poster reputation -- Computational linguistics; G02 -- G12; Investments -- Psychological aspects -- Periodicals; Investments -- Decision making -- Periodicals; 332.605
• DOI: 10.1108/RBF-05-2020-0117
• Languages: English
• ISSNs: 1940-5979

Abstract

Abstract : Purpose: The authors investigate the extent to which online talk can influence contemporaneous and future stock trading, especially when market news is unpresented. Design/methodology/approach: The authors propose an improved sentiment formula incorporating online hype, neutral sentiment and poster reputation. In addition, they conduct event study, OLS regression analyses and probit models. Findings: First, investors tend to be more talkative in relation to firms that are (1) smaller size, (2) more growth-like, (3) with lower prices and higher short interests and (4) of higher beta. Second, the bullish tone of investors positively affects the abnormal returns of small-capitalization stocks. However, online talk has little impact on large-capitalization stocks, except that more postings boost trading liquidity. Third, online talk predicts the presence of future news regardless of firm size, with stronger predictive power found for small-capitalization stocks. Practical implications: It is of interest to practitioners and researchers to study online talk so as to better understand the trading psychology of retail investors and the effects on the stock market. Furthermore, policymakers are interested in tracking activities on stock message boards in order to prevent security fraud and protect investors' interests. Originality/value: The results are robust and suggest that online talk has significant impacts on stock trading exploiting an information asymmetry. This study of stock message board posting activities helps researchers to understand whether message contents contain valuable and unique content compared with information available via more traditional media channels.