New study reveals GPT-4 outperforms humans in financial forecasting

OpenAI’s GPT-4 has emerged as a superior financial analyst compared to humans, as per a recent study conducted by the Booth School of Business at the University of Chicago. The study revealed that GPT-4 excelled in analyzing financial statements and making accurate predictions based on the data, even outperforming human analysts in certain scenarios.

The key takeaway from the study was that GPT-4’s ability to predict earnings changes was notably better than that of human analysts, especially when faced with incomplete information. By utilizing “chain-of-thought” prompts, the large language model was able to identify trends in financial statements, calculate various financial ratios, and make informed predictions about future earnings results.

The study highlighted that GPT-4 achieved an impressive accuracy rate of 60% in its predictions, significantly higher than the accuracy range of human analysts, which typically hovered around the low 50% mark. This suggests that GPT-4’s proficiency in recognizing financial patterns and business concepts with limited information could revolutionize financial decision-making processes in the future.

Moreover, the study demonstrated that trading strategies based on GPT-4’s forecasts yielded more profitable results than the stock market, with higher share ratios and alpha. The long-short strategy derived from GPT-4’s financial insights consistently outperformed the market, generating significant alphas and Sharpe ratios.

Overall, the study’s findings underscore the potential of advanced AI technologies like GPT-4 to disrupt and enhance the financial services industry, offering more accurate and profitable insights for investors and businesses alike.