Paper: arXiv 2311.06273

Authors: Ummara Mumtaz, Summaya Mumtaz

Abstract

The rise of ChatGPT has brought a notable shift to the AI sector, with its exceptional conversational skills and deep grasp of language. Recognizing its value across different areas, our study investigates ChatGPT’s capacity to predict stock market movements using only social media tweets and sentiment analysis. We aim to see if ChatGPT can tap into the vast sentiment data on platforms like Twitter to offer insightful predictions about stock trends. We focus on determining if a tweet has a positive, negative, or neutral effect on two big tech giants Microsoft and Google’s stock value. Our findings highlight a positive link between ChatGPT’s evaluations and the following days stock results for both tech companies. This research enriches our view on ChatGPT’s adaptability and emphasizes the growing importance of AI in shaping financial market forecasts.

Complexity vs Empirical Score

  • Math Complexity: 3.0/10
  • Empirical Rigor: 4.0/10
  • Quadrant: Philosophers — conceptual discussion, lower math and empirical scores

Why this score: The paper relies on prompt engineering and sentiment analysis using a pre-trained model (ChatGPT) with minimal custom mathematical modeling or statistical derivations. The empirical rigor is limited due to the use of a small, non-curated dataset, lack of robust backtesting, and minimal control for noise and market microstructure.

Research Flowchart

  flowchart TD
  A["Research Goal: Can ChatGPT predict<br>stock trends using Twitter sentiment?"]
  A --> B["Data Input:<br>Twitter Tweets"]
  B --> C["Computational Process:<br>ChatGPT Sentiment Analysis"]
  C --> D["Target: Microsoft & Google Stocks"]
  D --> E["Key Finding: Positive Correlation<br>between Sentiment & Next-Day Returns"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes using ChatGPT with zero-shot learning on Twitter sentiment to predict stock market trends for Microsoft and Google.

Source sentence
In this study, we explore the potential of using ChatGPT in predicting stock market trends solely based on Twitter sentiment analysis and by employing the zero-shot learning strategy.

Assumes 5

  • Minimal data cleaning (only URL removal) is sufficient for sentiment analysis. data
    Source sentence
    We performed minimal data cleaning and only removed URL present in the tweets.
  • Tweets mentioning Microsoft or Google contain relevant sentiment for those stocks. data
    Source sentence
    Tweets were specifically sought that mentioned or related to three distinct terms: 'ChatGPT', 'Microsoft', and 'Google'.
  • ChatGPT can predict stock trends without fine‑tuning, using zero‑shot learning. model
    Source sentence
    Considering ChatGPT was not exclusively fine-tuned for stock market predictions and employing a zero-shot learning strategy, its performance is notably superior to a model making random predictions.
  • The relationship between tweet sentiment and next‑day stock returns is stable over the sample period. market
    Source sentence
    Our findings highlight a positive link between ChatGPT's evaluations and the following days stock results for both tech companies.
Show 1 more
  • Trading can be evaluated purely on prediction accuracy without accounting for transaction costs. execution
    Source sentence
    The results indicate that ChatGPT achieved an accuracy of 70% for Microsoft and 63.88% for Google.

Data

  • 500k ChatGPT-related Tweets Jan-Mar 2023 · Kaggle alternative public
    2023-01-04 to 2023-03-29 · Tweets containing 'gpt', 'ChatGPT', 'Microsoft', or 'Google'
    Source sentence
    We utilized a publicly available dataset at Kaggle named “500k ChatGPT-related Tweets Jan-Mar 2023”
  • NASDAQ stock price data for Google and Microsoft · NASDAQ website market-prices public
    2023-01-04 to 2023-02-28 · Google and Microsoft stocks
    Source sentence
    Stock performance data for Google and Microsoft was procured from the NASDAQ website.

Recognised sources: Twitter / X Kaggle

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • ChatGPT correctly predicted Microsoft’s daily stock trend on 26 out of 37 days, yielding 70% accuracy. predictive-accuracy
    Source sentence
    For Microsoft, out of a 37-day period, ChatGPT's predictions aligned with the actual outcomes on 26 occasions. Thus, getting an accuracy of 70% for Microsoft stock market trend prediction.
  • ChatGPT correctly predicted Google’s daily stock trend on 23 out of 36 days, yielding roughly 63.88% accuracy. predictive-accuracy
    Source sentence
    out of a 36-day timeframe, the model made accurate forecasts on 23 days, resulting in an accuracy of roughly 63.88%.
  • A positive link was observed between ChatGPT’s sentiment evaluations and the next‑day stock results for both companies. performance
    Source sentence
    Our findings highlight a positive link between ChatGPT's evaluations and the following days stock results for both tech companies.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 24 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 1 paper in the archive (1 by independent authors)

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.