Chatgpt, Twitter Sentiment and Bitcoin Return

38 Pages Posted: 9 Nov 2023

See all articles by Binh Nguyen Thanh

Binh Nguyen Thanh

RMIT University Vietnam

Anh Tuan Nguyen

Royal Melbourne Institute of Technolog (RMIT University) - RMIT Vietnam

Thanh Tuan Chu

RMIT University Vietnam

Son Ha

Lecturer

Abstract

Our research augments the expanding body of literature concerning the capabilityof prevalent Generative AI tools in supporting financial professionals. Weintroduce a framework to leverage ChatGPT for assessing market sentimentthrough the analysis of social media data. We use the framework to constructmarket sentiment indicators based on Twitter tweets and use those indicators toexplain Bitcoin return. Our analysis uncovers that sentiment indicators craftedwith ChatGPT3.5 significantly affect Bitcoin returns, even when accounting fora broad array of control variables and other pre-established sentiment indicators.These insights imply that ChatGPT3.5 could empower financial professionalsto discover sentiment information from Twitter tweets that were overlooked bypreviously introduced sentiment indicators concerning Bitcoin.

Keywords: ChatGPT, Generative AI, Twitter sentiment, Bitcoin

Suggested Citation

Nguyen Thanh, Binh and Nguyen, Anh Tuan and Chu, Thanh Tuan and Ha, Son, Chatgpt, Twitter Sentiment and Bitcoin Return. Available at SSRN: https://ssrn.com/abstract=4628097 or http://dx.doi.org/10.2139/ssrn.4628097

Binh Nguyen Thanh

RMIT University Vietnam ( email )

The Business School, RMIT Vietnam
Ho Chi Minh City, 00000
Vietnam

Anh Tuan Nguyen (Contact Author)

Royal Melbourne Institute of Technolog (RMIT University) - RMIT Vietnam ( email )

Vietnam

Thanh Tuan Chu

RMIT University Vietnam ( email )

Son Ha

Lecturer ( email )

Vietnam

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

Paper statistics

Downloads
370
Abstract Views
1,274
Rank
174,716
PlumX Metrics