Being an emotionally unaffected investor: evidence from Bitcoin
Being an emotionally unaffected investor: evidence from Bitcoin
As one of the most prominent cryptocurrencies, Bitcoin has been at the forefront of a major revolution in the financial and technological sectors. This study utilizes data from social media to extract the emotional tendencies of investors in the Bitcoin market and analyze differences in investor behavior under various emotional features. We find that when investors exhibit reluctance (such as Sadness and Fear) to buy Bitcoin, it is the opportune moment to invest and achieve returns higher than expected. Conversely, when the emotional tone of investors becomes positive (such as Joy and Love), indicating a tendency to invest, we choose to avoid investing. Our research has also revealed that such emotional cues can assist in better predicting returns in the Bitcoin market. Analyzing market emotions contributes to a deeper understanding of market fluctuations and investor behavior. Our findings help stakeholders recognize the role of subjective emotions in the market and provide them with prudent investment advice: avoid relying excessively on the feelings of others, as this may trigger investment losses
BERT, Bitcoin, Investor Behavior, Large Language Models, Machine Learning, Return Prediction, Textual Analysis, investor behavior, return prediction, machine learning, textual analysis, large language models, bitcoin
1471-1485
Ren, Xiaohang
644d6dde-0155-4872-9488-336acf86dba2
Jiang, Wenting
a5b1d76a-762e-4bd4-af52-e60e14d49694
Duan, Kun
09eaaadf-6d46-406b-9b51-11edc63bea49
Mishra, Tapas
218ef618-6b3e-471b-a686-15460da145e0
25 March 2025
Ren, Xiaohang
644d6dde-0155-4872-9488-336acf86dba2
Jiang, Wenting
a5b1d76a-762e-4bd4-af52-e60e14d49694
Duan, Kun
09eaaadf-6d46-406b-9b51-11edc63bea49
Mishra, Tapas
218ef618-6b3e-471b-a686-15460da145e0
Ren, Xiaohang, Jiang, Wenting, Duan, Kun and Mishra, Tapas
(2025)
Being an emotionally unaffected investor: evidence from Bitcoin.
IEEE Transactions on Engineering Management, 72, .
(doi:10.1109/TEM.2025.3554567).
Abstract
As one of the most prominent cryptocurrencies, Bitcoin has been at the forefront of a major revolution in the financial and technological sectors. This study utilizes data from social media to extract the emotional tendencies of investors in the Bitcoin market and analyze differences in investor behavior under various emotional features. We find that when investors exhibit reluctance (such as Sadness and Fear) to buy Bitcoin, it is the opportune moment to invest and achieve returns higher than expected. Conversely, when the emotional tone of investors becomes positive (such as Joy and Love), indicating a tendency to invest, we choose to avoid investing. Our research has also revealed that such emotional cues can assist in better predicting returns in the Bitcoin market. Analyzing market emotions contributes to a deeper understanding of market fluctuations and investor behavior. Our findings help stakeholders recognize the role of subjective emotions in the market and provide them with prudent investment advice: avoid relying excessively on the feelings of others, as this may trigger investment losses
Text
draft_Proof_hi
- Accepted Manuscript
More information
Accepted/In Press date: 13 March 2025
e-pub ahead of print date: 25 March 2025
Published date: 25 March 2025
Keywords:
BERT, Bitcoin, Investor Behavior, Large Language Models, Machine Learning, Return Prediction, Textual Analysis, investor behavior, return prediction, machine learning, textual analysis, large language models, bitcoin
Identifiers
Local EPrints ID: 499955
URI: http://eprints.soton.ac.uk/id/eprint/499955
ISSN: 0018-9391
PURE UUID: f87bfa40-bc37-43c8-b650-4fdf7f8748cc
Catalogue record
Date deposited: 09 Apr 2025 18:45
Last modified: 27 Aug 2025 01:49
Export record
Altmetrics
Contributors
Author:
Xiaohang Ren
Author:
Wenting Jiang
Author:
Kun Duan
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
