GPT-4 Outperforms Human Financial Analysts In Predicting Earnings Growth, Say Researchers

IBL News | New York

Researchers from the University of Chicago found that OpenAI’s GPT-4 outperforms human financial analysts in predicting future earnings growth.

They reflected their conclusions in a working paper titled “Financial Statement Analysis with Large Language Models.

The findings could have significant implications for financial analysis and decision-making, according to VentureBeat.

The researchers used “chain-of-thought” prompts to emulate the analytical process of a financial analyst, identifying trends, computing ratios, and synthesizing the information to form a prediction.

They achieved a 60% accuracy in predicting the direction of future earnings, notably higher than the 53-57% range of human analyst forecasts.

“Our results suggest that LLMs may take a central role in decision-making,” the authors concluded.