Forecasting Short-Term Stock Returns Using Large Language Models: A Comparative Analysis with Human Analysts
Abstract
This study aims to determine whether artificial intelligence can accurately predict short-term returns of large-cap stocks following large announcements or significant news. ChatGPT and Gemini were asked to predict the performance of certain stocks over a week period and over a month period. Additionally, professional analyst data and predictions were compared to AI’s results. Results showed that both AI and analysts performed poorly, with AI slightly beating the analysts. These findings suggest that short-term investing as a whole may not be possible to predict, but also that access to past market patterns is helpful to create a forecast. Future research should examine smaller stocks, longer periods of time, and the use of AI with human judgment.
Published
Issue
Section
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Authors who publish with this journal agree to the following terms: RAIS Journal of Social Sciences is given by the author the right of the first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal. Authors retain copyright. If the author cites from his own article published in RAIS Journal of Social Sciences, then he is encouraged to cite the name of the RAIS Journal of Social Sciences, volume, and page. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access). This journal provides immediate open access to its content, in this way, we make research freely available to the public and support a greater global exchange of knowledge.
PRIVACY STATEMENT
The names and email addresses entered in this journal site will be used exclusively for the stated purposes of this journal and will not be made available for any other purpose or to any other party.
