ChatGPT may be able to predict stock movements: finance professor

ChatGPT could be the next stock forecaster, according to this finance professor

Alejandro Lopez-Lira, a finance professor at the University of Florida, says that large language models may be useful when forecasting stock prices.

He used ChatGPT to parse news headlines for whether they’re good or bad for a stock, and found that ChatGPT’s ability to predict the direction of the next day’s returns were much better than random, he said in a recent unreviewed paper.

The experiment strikes at the heart of the promise around state-of-the-art artificial intelligence: With bigger computers and better datasets — like those powering ChatGPT — these AI models may display “emergent abilities,” or capabilities that weren’t originally planned when they were built.

If ChatGPT could display the emergent ability to understand headlines from financial news and how they might impact stock prices, it could put high-paying jobs in the financial industry at risk. About 35% of financial jobs are at risk of being automated by AI, Goldman Sachs estimated in a March 26 note.

“The fact that ChatGPT is understanding information meant for humans almost guarantees if the market doesn’t respond perfectly, that there will be return predictability,” said Lopez-Lira.

A trader works on the floor of the New York Stock Exchange.

Jason Decrow

But the specifics of the experiment also show how so-called “large language models” are from being able to do many financial tasks.

For example, the experiment didn’t include target prices, or have the model do any math at all. In fact, ChatGPT-style technology often makes numbers up, as Microsoft learned in a public demo earlier this year. Sentiment analysis of headlines is also well understood as a trading strategy, with proprietary datasets already in existence.

Lopez-Lira said he was surprised by the results, adding they suggest that sophisticated investors aren’t using ChatGPT-style machine learning in their trading strategies yet.

“On the regulatory side, if we have computers just reading the headlines, the headlines will matter more, and we can see if everyone should have access to machines such as GPT,” said Lopez-Lira. “Second, it’s certainly going to have some implications for the employment of financial analyst landscape. The question is, do I want to pay analysts? Or can I just put textual information in a model?”

How the experiment worked

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 started in October 2022 — after the data cutoff date for ChatGPT, meaning that the engine hadn’t seen or used those headlines in training.

Then, they fed the headlines into ChatGPT 3.5 along with the following prompt:

“Forget all your previous instructions. Pretend you are a financial expert. You are a financial expert with stock recommendation experience. Answer “YES” if good news, “NO” if bad news, or “UNKNOWN” if uncertain in the first line. Then elaborate with one short and concise sentence on the next line.”

Then they looked at the stocks’ return during the following trading day.

Ultimately, Lopez-Lira found that the model did better in almost all cases when informed by a news headline. Specifically, he found a less than 1% chance the model would do as well as picking the next day’s move at random, versus when it was informed by a news headline.

ChatGPT also beat commercial datasets with human sentiment scores. One example in the paper showed a headline about a company settling litigation and paying a fine, which had a negative sentiment, but the ChatGPT response correctly reasoned it was actually good news, according to the researchers.

Lopez-Lira told CNBC that hedge funds had reached out to him to learn more about his research. He also said it wouldn’t surprise him if ChatGPT’s ability to predict stock moves decreased in the coming months as institutions started integrating this technology.

That’s because the experiment only looked at stock prices during the next trading day, while most people would expect the market could have already priced the news in seconds after it became public.

“As more and more people use these types of tools, the market is going to become more efficient, so you would expect return predictability to decline,” said Lopez-Lira. “So my guess is, if I run this exercise, in the next five years, by the year five, there will be zero return predictability.” powergaming88 powergaming88 powergaming88 powergaming88 powergaming88 powergaming88 powergaming88 powergaming88 slot zeus nolimit city sv388 slot jepang scatter hitam slot thailand link slot thailand 777 slot powergaming88 slot luar negeri slot gacor maxwin slot gacor hari ini slot gacor terbaru mahjong ways 2 slot thailand jp slot slot bni slot bri slot bsi slot danamon slot gopay slot server korea slot malaysia slot amerika slot rusia fan tan monopoly live lightning roulette mega ball sweet bonanza candyland mega sic bo mega baccarat andar bahar sbobet88 ktm303 slot dubai slot qris powergaming88 Slot Thailand Slot Zeus Pg Soft Starlight Princess Slot Gacor Maxwin Joker123 Sv388 Slot Thailand Slot Thailand Slot Qris slot thailand joker123 mahjong ways 2 888slot sweet bonanza slot jepang slot thailand slot mahjong ways slot qris slot deposit dana slot dana nolimit city slot thailand sv388 slot888 slot kamboja Slot Thailand Starlight Princess Mahjong Ways 2 Slot Maxwin slot thailand akun pro kamboja slot filipina joker123 slot zeus slot88 slot thailand slot mahjong slot megaways NEW SLOT POLA GACOR Slot Server Luar jp slot slot ganas crazy time live slot qris slot akurat slot bca slot bni qris slot zeus server thailand server hongkong slot ovo slot gopay slot linkaja joker123 nolimit city sv388 starlight princess mahjong ways pg slot slot zeus sv388 idn slot slot server thailand slot thailand ladangtoto nolimit city koi gate sicbo online roulette online lice casino live22 baccarat online nolimit city slot ovo slot gopay pragmati play starlight princess slot pg soft slot mahjong slot zeus sv388 digmaan slot mania slot zeus slot thailand server thailand ying77 slot thailand judi bola live casino sv388 pg soft slot dana mahjong ways slot mania slot x500 ladangtoto slot server kamboja digmaan digmaan slot88 scv388 sabung ayam online slot live casino sicbo online live casino online slot thailand ladangtoto sv388 ladangtoto powergaming88 66kbet powergaming88 slot thailand ladangtoto slot thailand server thailand slot thailand slot thailand slot thailand slot gacor ladangtoto2 ying77 bet server thailand slot thailand ying77 slot88 ladangtoto ladangtoto slot maxwin sv388 sv388 slot gacor 66k bet slot dana sabung ayam inline slot mahjong slot toto slot server thailand slot gacor slot server thailand slot gacor slot thailand situs slot wwg slot slot daftar judi slot online indonesia slot gacor slot mawin slot gacor slot thailand slot gacor slot gacor demo slot pragmatic sv388 slot88 pgsoft Daftar Judi Slot Online Indonesia Paling Gacor di Indonesia Daftar Situs Judi Slot Online SLOT88 lucky neko slot gacor slot88 slot kakek slot maxwin slot thailand Slot Mahjong Ways 2 Mudah Jackpot Paus di PG Soft Hari ini 2023 Lucky Neko Situs Slot Gacor Pgsoft Hari Ini SITUS SABUNG AYAM SV388 ONLINE DAFTAR LINK LOGIN APK AGEN SV388
Scroll to Top