Wisdom or Whims? Decoding Investor Trading Strategies with Large Language Models

64 Pages Posted: 25 Jun 2024

See all articles by Shuaiyu Chen

Shuaiyu Chen

Purdue University - Mitchell E. Daniels, Jr. School of Business

Lin Peng

City University of New York, Baruch College - Zicklin School of Business - Department of Economics and Finance

Dexin Zhou

City University of New York, Baruch College - Zicklin School of Business - Department of Economics and Finance

Date Written: June 19, 2024

Abstract

This paper analyzes the trading strategies used by retail investors. By applying large-language models (LLMs) to over 77 million messages posted by nearly 800,000 users on a popular social media platform, we obtain sharp inferences about investor strategies. Retail investors tend to rely more on technical analysis (TA) strategies when fundamental news is scarce, and TA sentiment negatively predicts future returns, is associated with less informative retail order flows, and a higher likelihood of Robinhood herding episodes. In contrast, non-TA messages, especially those describing fundamental analysis, exhibit sentiment that is informative of future returns and tend to attenuate herding. Furthermore, the state-of-the-art AI signal of Jiang et al. (2023) generates significant profits when trading against TA sentiment but is no longer profitable when the signal coincides with TA sentiment. The findings suggest that a key source of AI signal profitability is retail investors' tendency to misinterpret technical signals. The evidence provides insights into the diverse investment approaches traders use, the role of social media, and the interactions between different players in the age of AI-powered trading.

Keywords: Social media, Retail investors, Large-language models, AI, Technical analysis, Fundamental analysis

Suggested Citation

Chen, Shuaiyu and Peng, Lin and Zhou, Dexin, Wisdom or Whims? Decoding Investor Trading Strategies with Large Language Models (June 19, 2024). Available at SSRN: https://ssrn.com/abstract=4867401 or http://dx.doi.org/10.2139/ssrn.4867401

Shuaiyu Chen (Contact Author)

Purdue University - Mitchell E. Daniels, Jr. School of Business ( email )

1310 Krannert Building
West Lafayette, IN 47907-1310
United States
5853198838 (Phone)
47906-1744 (Fax)

Lin Peng

City University of New York, Baruch College - Zicklin School of Business - Department of Economics and Finance ( email )

17 Lexington Avenue
New York, NY 10010
United States

Dexin Zhou

City University of New York, Baruch College - Zicklin School of Business - Department of Economics and Finance ( email )

55 Lexington Avenue
New York, NY 10010
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
66
Abstract Views
311
Rank
638,170
PlumX Metrics