Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

The Influence of Trump’s Social Media Sentiment on Bitcoin Price Returns

Gouveia, Joaquim LU and Johansson, Oliver LU (2025) In Bachelor Thesis in Economics, Lund University NEKH02 20251
Department of Economics
Abstract
This thesis examines the impact of Donald Trump’s social media sentiment
on Bitcoin price returns, comparing two periods with distinct political con-
texts: during his presidency in 2019 and his re-election campaign in 2024.
Employing VADER, a sentiment analysis tool optimized for short social me-
dia texts, the study analyzes the sentiment of Trump’s tweets and Truth
Social posts. Using OLS regression, it models their effect on Bitcoin’s daily
returns to assess whether sentiment influences prices and whether this effect
varies between presidency and re-election campaign. The results indicate
a noticeable divergence between the two time periods. In 2019, tweet sen-
timent significantly predicted Bitcoin daily price changes,... (More)
This thesis examines the impact of Donald Trump’s social media sentiment
on Bitcoin price returns, comparing two periods with distinct political con-
texts: during his presidency in 2019 and his re-election campaign in 2024.
Employing VADER, a sentiment analysis tool optimized for short social me-
dia texts, the study analyzes the sentiment of Trump’s tweets and Truth
Social posts. Using OLS regression, it models their effect on Bitcoin’s daily
returns to assess whether sentiment influences prices and whether this effect
varies between presidency and re-election campaign. The results indicate
a noticeable divergence between the two time periods. In 2019, tweet sen-
timent significantly predicted Bitcoin daily price changes, suggesting that
retail investors were highly responsive to Trump’s presidential communica-
tions. In contrast, Trump’s 2024 campaign posts on Truth Social showed
no significant impact on Bitcoin prices, likely due to the cryptocurrency’s
increased market maturity and the platform’s limited reach compared to
Twitter. Consequently, a Chow test was conducted, revealing a potential
structural break, suggesting a fundamental shift in how political sentiment
affects market behavior across these periods. These findings carry impor-
tant theoretical implications. They challenge semi-strong Efficient Market
Hypothesis in 2019 by highlighting behavioral biases, while supporting it in
2024, indicating greater market efficiency. Despite limitations such as short
sample periods and noisy sentiment measures, the study underscores the
evolving role of political communication in shaping sentiment-driven mar-
kets, offering insights for both finance and behavioral economics. (Less)
Please use this url to cite or link to this publication:
author
Gouveia, Joaquim LU and Johansson, Oliver LU
supervisor
organization
course
NEKH02 20251
year
type
M2 - Bachelor Degree
subject
keywords
Bitcoin, Donald Trump, Sentiment Analysis, Efficient Market, Hypothesis, Behavioral Finance.
publication/series
Bachelor Thesis in Economics, Lund University
language
English
id
9220074
date added to LUP
2026-02-04 08:29:27
date last changed
2026-02-04 08:29:27
@misc{9220074,
  abstract     = {{This thesis examines the impact of Donald Trump’s social media sentiment
on Bitcoin price returns, comparing two periods with distinct political con-
texts: during his presidency in 2019 and his re-election campaign in 2024.
Employing VADER, a sentiment analysis tool optimized for short social me-
dia texts, the study analyzes the sentiment of Trump’s tweets and Truth
Social posts. Using OLS regression, it models their effect on Bitcoin’s daily
returns to assess whether sentiment influences prices and whether this effect
varies between presidency and re-election campaign. The results indicate
a noticeable divergence between the two time periods. In 2019, tweet sen-
timent significantly predicted Bitcoin daily price changes, suggesting that
retail investors were highly responsive to Trump’s presidential communica-
tions. In contrast, Trump’s 2024 campaign posts on Truth Social showed
no significant impact on Bitcoin prices, likely due to the cryptocurrency’s
increased market maturity and the platform’s limited reach compared to
Twitter. Consequently, a Chow test was conducted, revealing a potential
structural break, suggesting a fundamental shift in how political sentiment
affects market behavior across these periods. These findings carry impor-
tant theoretical implications. They challenge semi-strong Efficient Market
Hypothesis in 2019 by highlighting behavioral biases, while supporting it in
2024, indicating greater market efficiency. Despite limitations such as short
sample periods and noisy sentiment measures, the study underscores the
evolving role of political communication in shaping sentiment-driven mar-
kets, offering insights for both finance and behavioral economics.}},
  author       = {{Gouveia, Joaquim and Johansson, Oliver}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{Bachelor Thesis in Economics, Lund University}},
  title        = {{The Influence of Trump’s Social Media Sentiment on Bitcoin Price Returns}},
  year         = {{2025}},
}