Is ChatGPT a Good Stock Picker?

This research paper analyzes the ability of ChatGPT to classify sentiment on corporate earnings calls.  We compare the performance of ChatGPT to popular NLP approaches in the investment industry: FinBERT, Loughran McDonald, and Alexandria's machine learning approach. 

We find:

  • Using Sentiment for Earnings Calls can generate alpha and lead to outperformance not explained by traditional risk and return factors.
  • High correlations of ChatGPT sentiment classification with FinBERT and the Loughran McDonald dictionary.
  • Large performance differences in return prediction for the various approaches

*Sentiment derived from FactSet XML Transcripts

Request Research

This site uses cookies, if you continue without changing your settings, we'll assume that you are happy to receive all cookies. Click here to learn more about cookies.