Impact of Analytical AI on Equity Volatility

Abstract

This study specifically examines how the news of publicly traded companies implementing analytical AI can affect investor sentiment and behaviour, as measured through short term changes in return volatility of stocks and average returns. Using an event study methodology, this study analyzes three publicly traded companies—Walmart, Hilton, and Amazon—each with a specific and dateable disclosure of analytical AI implementation. For each company, daily returns and return volatility were compared across pre and post event windows of 1, 5, and 7 trading days, using independent two sample t-tests and Levene’s tests of equality of variances. Ordinary least squares regressions were also estimated to assess whether the event date predicted returns beyond what the volatility would suggest. Contrary to the study’s hypothesis that AI implementation disclosures would be associated with increased short term volatility, no company exhibited a statistically significant change in volatility at any window length. However, Hilton demonstrated a statistically significant decline in average returns at both the 5 day and 7 day windows, a result not detected in the regression analysis, suggesting that any sentiment effect present may be short-lived and concentrated immediately around the event. Given the small sample size and the number of statistical comparisons conducted, these findings are best interpreted as exploratory into potential effects of news about AI implementation by certain corporations rather than conclusive evidence of a broader relationship between analytical AI disclosures and investor sentiment.