ChatGPT improves news-based stock predictions than current models: study

  • Florida researchers asked ChatGPT to analyze headline sentiment to predict resulting stock movements.
  • They said their study “demonstrates the value of ChatGPT in predicting stock returns.”
  • But ChatGPT still has flaws, and researchers advise investors not to rely on it for their investment decisions.

ChatGPT can’t see the future, but it already has value for investors looking to predict future stock market movements.

It is according to a new research paper published Monday in the Social Science Research Network by two University of Florida finance professors.

The researchers fed the chatbot more than 50,000 company news headlines dating back to October 2021 and asked it to determine whether the headlines were good, bad or irrelevant to company stock prices. They used this sentiment analysis to calculate a numerical “ChatGPT score” and analyzed whether these scores were predictive of the company’s stock market performance the next day.

The researchers found a statistically significant positive correlation between these scores and next-day stock market performance for the companies they analyzed. Companies with higher scores tend to see better returns than those with lower scores.

The study also found that ChatGPT outperformed other “traditional sentiment analysis methods” that also use headline and social media data to predict stock movements – although the researchers admit they didn’t. tested each of these methods in this study.

“In short, our study demonstrates the value of ChatGPT in predicting stock market returns,” the researchers wrote.

They added, “Our results suggest that integrating advanced language models into the investment decision-making process can produce more accurate predictions and improve the performance of quantitative trading strategies.”

When the researchers compared the performance of ChatGPT with traditional methods, they found that these other models added no predictive power to their ChatGPT-derived sentiment scores.

“Once you use ChatGPT, using the other sentiment metric is not useful for predicting,” Alejandro Lopez-Lira, co-author of the paper, told Insider.

Investors should always refrain from relying solely on ChatGPT

Lopez-Lira said that in the future, AI tools like ChatGPT could be used to improve stock market efficiency by integrating news more quickly into stock prices. He also said these tools could replace some investment analysts.

From ChatGPT was deployed last November, users have tested its limits. They asked for it what stocks to invest inused it for make money from side scamsand even asked him to help start a business. As much as the chatbot surprised, as many of its limits also appeared.

Lopez-Lira said investors should “use caution and not rely solely on ChatGPT or similar AI models,” in part because there are several areas where AI needs to improve. regarding stock market forecasts.

First of all, ChatGPT is not “connected to the Internet” and therefore cannot access the latest information available.

“ChatGPT does not have access to any recent data beyond its September 2021 training deadline,” he said. “This limitation means that the AI ​​model may not be aware of more recent market trends, news, or developments that could significantly impact stock prices and investment decisions.”

ChatGPT also struggles with large texts and “calculating numbers”.

“It cannot process large amounts of digital data, such as accounting data from companies,” he said.

Addressing these concerns, Lopez-Lira said, could “significantly increase the predictive capabilities” of the chatbot.

“As the field of AI-driven finance continues to develop, the insights gained from this research can help guide the development of more accurate, efficient, and accountable models that improve the performance of financial decision-making processes.” , wrote the researchers in the paper’s conclusion line.

Leave a Comment