HomeArticlesServicePriceAbout

The Usefulness of ChatGPT for Textual Analysis of Annual Reports

This study examines the use of ChatGPT to analyze the textual complexity and sentiment of UK annual reports and investigates its impact on investment decisions. The findings indicate that the sentiment and complexity scores generated by ChatGPT contain economically significant information, influencing both stock price reactions to annual report releases and future profitability. Additionally, these indicators capture investor belief heterogeneity, providing insights into how investors process the textual content of annual reports. Ultimately, the study suggests that investors can utilize AI models like ChatGPT to analyze textual characteristics of annual reports and enhance investment decisions.

Key Research Methods

• Data: Annual report PDFs of FTSE100 companies from 2015 to 2022 were collected.

• ChatGPT-based Text Analysis: ChatGPT-4 was used to calculate sentiment and complexity scores for key sections of the reports (Chairman’s statement, performance highlights, corporate governance, and ESG disclosures).

• Index Construction: Sentiment and complexity scores were processed using principal component analysis to create a Sentiment Index and a Complexity Index.

• Regression Analysis: Sentiment and complexity indices were used as independent variables, while stock price reactions, future profitability, and investor belief heterogeneity were set as dependent variables.

• Robustness Checks: Fixed effects for firms and years were included, and standard errors were clustered to ensure the stability of the results.

Key Findings

• Useful Information: The sentiment and complexity indices generated by ChatGPT contain valuable information that influences stock price reactions and future profitability.

• Impact of Sentiment Index: A higher sentiment index is associated with positive stock price reactions and increased future ROA (Return on Assets).

• Impact of Complexity Index: A higher complexity index is linked to negative stock price reactions and lower future ROA growth.

• Investor Belief Heterogeneity: The sentiment index negatively correlates with stock turnover and return volatility, suggesting that positive sentiment reduces differences in investor opinions. Conversely, the complexity index positively correlates with turnover, indicating that higher complexity increases disagreement among investors.

Conclusion

This study demonstrates that ChatGPT effectively analyzes textual characteristics of annual reports and extracts valuable information for investment decisions. Investors can leverage ChatGPT to assess the sentiment embedded in corporate disclosures and refine their investment strategies.

[Compliance Note]

• All posts by Sellsmart are for informational purposes only. Final investment decisions should be made with careful judgment and at the investor’s own risk.

• The content of this post may be inaccurate, and any profits or losses resulting from trades are solely the responsibility of the investor.

• Core16 may hold positions in the stocks mentioned in this post and may buy or sell them at any time.