Can ChatGPT predict Chinese equity premiums?

Feng Ma, Zhichong Lyu and Haibo Li

Finance Research Letters, 2024, vol. 65, issue C

Abstract: Leveraging over 1.86 million news headlines, we examine the capability of ChatGPT-3.5, a large language model (LLM), to predict equity risk premiums in the Chinese market. The predictive scores from ChatGPT not only positively and significantly forecast equity premiums but also markedly outperform the bag-of-words (BoW) method, demonstrating its superior capability to discern intricate market sentiments from extensive datasets. The consistent and reliable performance in both in-sample and out-of-sample tests underscores the effectiveness of ChatGPT and its potential to revolutionize financial forecasting. This study highlights the substantial value and innovative contribution of LLMs, such as ChatGPT, in enriching the precision and depth of financial market analysis.

Keywords: Large language model; ChatGPT; Chinese equity premium; Bag-of-words (search for similar items in EconPapers)
JEL-codes: C53 G12 G17 O33 (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S1544612324006615
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:finlet:v:65:y:2024:i:c:s1544612324006615

DOI: 10.1016/j.frl.2024.105631

Access Statistics for this article

Finance Research Letters is currently edited by R. Gençay

More articles in Finance Research Letters from Elsevier
Bibliographic data for series maintained by Catherine Liu ().