Market Reaction to Verbal Components of Earnings Press Releases: Event Study Using a Predictive Algorithm

Abstract

Similar to a classic-event study, this study examines market reaction to firmsa' earnings announcements. This study extends the examination to include a broad range of concurrent disclosure contained in earnings press releases: financial disclosure captured as accounting ratios; and verbal components of disclosure, both content and style, which are captured using elementary computer-based content analysis. Extending the analysis to such a broad range of concurrent disclosures requires a methodology designed to utilize a large number of predictor variables, and predictive data mining algorithms are specifically designed to do so. Therefore, this study employs a widely used data-mining algorithm—classification and regression trees (CART). Results of the study show that inclusion of predictor variables capturing verbal content and writing style of earnings-press releases results in more accurate predictions of market response.