Now, ChatGPT to predict stock prices by analysing news headlines? Florida Professor shows how
3 min read 12 May 2023, 02:12 PM ISTAlejandro Lopez-Lira used Open AI's chatbot ChatGPT to examine news headlines through 'sentimental analysis' for whether they are good or bad for a stock and found that the chatbot was able to predict the direction of the next day's returns
PremiumA finance professor at the University of Florida has claimed that large language models (LLM) - a type of artificial intelligence (AI) algorithm, may be useful in predicting stock market prices. Alejandro Lopez-Lira used Open AI's chatbot ChatGPT to examine news headlines through 'sentimental analysis' for whether they are good or bad for a stock and found that the chatbot was able to predict the direction of the next day's returns, according to his recently-published unreviewed paper.
As part of the study, the experiment conducted by Lopez-Lira and his partner Yuehua Tan is based on the emerging abilities of AI models, especially those powering ChatGPT. LLMs are more capable of understanding natural language and can process textual information to predict stock returns, However, the paper adds, ‘’these models are not explicitly trained for this purpose, one may expect that they offer little value in predicting stock market movements.''
‘’We use ChatGPT to indicate whether a given headline is good, bad, or irrelevant news for firms' stock prices. We then compute a numerical score and document a positive correlation between these "ChatGPT scores" and subsequent daily stock market returns. Further, ChatGPT outperforms traditional sentiment analysis methods,'' said the abstract of the paper.
‘’Our results suggest that incorporating advanced language models into the investment decision-making process can yield more accurate predictions and enhance the performance of quantitative trading strategies,'' it added.
Also Read: These jobs are most at risk due to ChatGPT, as per OpenAI study
How the experiment worked
Lopez-Lira and his partner Yuehua Tang looked at over 50,000 headlines from a data vendor about public shares listed on the Nasdaq, New York Stock Exchange and the American Stock Exchange. The sample period began in October 2021 (as ChatGPT’s training data is available only until September 2021) and ended in December 2022.
The headlines were fed into ChatGPT 3.5 along with the prompt and the stocks returns were assessed during the next trading day. A prompt is a short piece of text that provides context and instructions for ChatGPT to generate a response and are essential for enabling the chatbot to perform a language tasks, such as language translation, text summarization, generating human-like text, among others.
The Professors found that the model did better in nearly all cases when informed by a news headline. Specifically, they found a less than 1 per cent chance the model would do as well 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 paper, the superiority of ChatGPT in predicting stock market returns can be attributed to its advanced language understanding capabilities, which allow it to capture the nuances and subtleties within news headlines. This enables the model to generate more reliable sentiment scores, leading to better predictions of daily stock market returns. The experiment did not include target price or the model did not carry out mathematical task.
Also Read: Microsoft to launch new version of ChatGPT, will solve privacy concerns
What are the major findings of the study
The implementation of the results can potentially lead to a shift in the methods used for market prediction and investment decision-making. Here are the three major findings of the research conducted as part of the paper, which can lead to practical implications in the industry:
-The research can aid regulators and policymakers in understanding the potential benefits and risks associated with the increasing adoption of LLMs in financial markets. It can lead to discussions on regulatory frameworks that govern the use of AI in finance and lead to development of integrating LLMs into market operations.
-The study can benefit asset managers and institutional investors by providing empirical evidence on the efficacy of LLMs in predicting stock market returns. It can help professionals make more informed decisions about incorporating LLMs into their investment strategies, potentially leading to improved performance and reduced reliance on traditional, labor-intensive analysis methods.
-The research contributes to the broader academic discourse on artificial intelligence applications in finance. By exploring the capabilities of ChatGPT in predicting stock market returns, LLMs’ potential and limitations within the financial economics domain can be widely understood.
"Exciting news! Mint is now on WhatsApp Channels 🚀 Subscribe today by clicking the link and stay updated with the latest financial insights!" Click here!
