WEBINAR on The Effects of Bots on Market Reactions to Earnings Announcement Events
This one-hour webinar will present recent academic research that has been funded by CPA Ontario and the Schulich CPA Ontario Centre in Digital Financial Information.
Overview
Social media platforms such as Twitter have attracted millions – and for some platforms, billions – of people to tweet, pin, post, upload and share their latest ideas, thoughts, actions, and meals. There is ample evidence that, collectively, these social media users’ posts can influence the capital markets. Yet beyond people, social media platforms are attracting non-human bots. Starting with the proposition that algorithm-driven, non-human bots can affect the course of online discussion networks and thereby influence stock market reactions to new information, we use data analytics and machine learning tools to measure the level of bot activity in the 12.02 million Twitter messages discussing the stocks of the firms in the S&P 1,500 index in 2018. Each of these 1,500 stocks has a distinct Twitter discussion network, and we hypothesize that the intensity of the market reaction to an earnings announcement will be contingent on the extent of bot activity in that network. We test this hypothesis by examining the influence of bots on market reactions to 2018 quarterly earnings announcement events. Our findings corroborate our core hypothesis: in the presence of good earnings news, more extensive bot activity is associated with increased abnormal returns, while the opposite occurs with bad earnings news. We also show this effect is stronger the more bot tweets are shared by other Twitter users.
While being a largely "academic research"-focused presentation, CPA practitioners should be interested in the cutting-edge insights from this accounting research. Webinar participants will receive an official verified confirmation of participation after the webinar that can be used toward CPA professional development requirements.
Overview
Social media platforms such as Twitter have attracted millions – and for some platforms, billions – of people to tweet, pin, post, upload and share their latest ideas, thoughts, actions, and meals. There is ample evidence that, collectively, these social media users’ posts can influence the capital markets. Yet beyond people, social media platforms are attracting non-human bots. Starting with the proposition that algorithm-driven, non-human bots can affect the course of online discussion networks and thereby influence stock market reactions to new information, we use data analytics and machine learning tools to measure the level of bot activity in the 12.02 million Twitter messages discussing the stocks of the firms in the S&P 1,500 index in 2018. Each of these 1,500 stocks has a distinct Twitter discussion network, and we hypothesize that the intensity of the market reaction to an earnings announcement will be contingent on the extent of bot activity in that network. We test this hypothesis by examining the influence of bots on market reactions to 2018 quarterly earnings announcement events. Our findings corroborate our core hypothesis: in the presence of good earnings news, more extensive bot activity is associated with increased abnormal returns, while the opposite occurs with bad earnings news. We also show this effect is stronger the more bot tweets are shared by other Twitter users.
While being a largely "academic research"-focused presentation, CPA practitioners should be interested in the cutting-edge insights from this accounting research. Webinar participants will receive an official verified confirmation of participation after the webinar that can be used toward CPA professional development requirements.
