Date of Completion

Spring 5-1-2026

Thesis Advisor(s)

Lingling Wang

Honors Major

Finance

Disciplines

Behavioral Economics | Finance and Financial Management

Abstract

This paper examines whether increased attention to artificial intelligence (AI) influences herding behavior in financial markets. As AI has become a dominant theme in both financial markets and public discourse, understanding how investor attention affects behavior has become increasingly important. Herding behavior is measured using the Cross-Sectional Absolute Deviation (CSAD) methodology developed by Chang, Cheng, and Khorana (2000). The analysis uses daily data from December 2019 through November 2025 and focuses on the Global X Artificial Intelligence & Technology ETF (AIQ) as a proxy for the AI sector. Investor attention is measured using a composite Google Search Volume Index (SVI) constructed from multiple AI-related search terms.

The baseline results provide evidence of herding behavior within the AI sector, as return dispersion increases at a decreasing rate during periods of large market movements. However, the findings also indicate that herding weakens following the introduction of ChatGPT and during periods of elevated investor attention. In both cases, the nonlinear herding coefficient becomes less negative and loses statistical significance, suggesting that investors rely less on market-wide signals when information is more accessible and attention to AI is greater.

Overall, the results suggest that technological advancements and increased information availability may reduce behavioral biases and encourage more independent investment decision making. These findings contribute to the behavioral finance literature by demonstrating that herding behavior varies across informational environments.

Accessibility Requirements

1

Download

Share

COinS