ChatGPT may be useful in predicting stock prices.???
According to Alejandro Lopez-Lira, a finance professor at the University of Florida, large language models like ChatGPT may be useful in predicting stock prices. He conducted an experiment where he fed ChatGPT news headlines about public stocks and asked it to predict whether the news was good or bad for the stock. The results showed that ChatGPT's ability to predict the direction of the next day's returns was much better than random chance.
Lopez-Lira believes that if ChatGPT can understand financial news and predict stock prices, it could put high-paying jobs in the financial industry at risk. However, the experiment also showed that large language models are still far from being able to perform many finance tasks. The experiment did not include target prices, nor did the model perform any math.
Lopez-Lira was surprised by the results and suggested that sophisticated investors are not yet using ChatGPT-style machine learning in their trading strategies. He also noted that as more people use these tools, the markets will become more efficient, and return predictability will decline.
In the experiment, Lopez-Lira and his partner Yuehua Tang looked at over 50,000 headlines from a data vendor about public stocks on the New York Stock Exchange, Nasdaq, and a small-cap exchange. They fed the headlines into ChatGPT along with a prompt asking it to predict whether the news was good or bad for the stock. They then looked at the stocks' return during the following trading day.
Overall, Lopez-Lira found that ChatGPT performed better when informed by a news headline than when picking the next day's move at random. Hedge funds have reportedly reached out to Lopez-Lira to learn more about his research.


留言
張貼留言