The author, however, warns that there is still imperfection in ChatGPT and that it is still far from performing specific calculations, setting target price of shares, among other abilities.
Hawkish or dovish, ChatGPT cracks it
But it’s not all down to news headlines. A second study, carried out by two researchers from the United States Federal Reserve (Fed), tested the ability of ChatGPT to decipher a monetary policy speech (Fedspeak).
The results showed that the chatbot can interpret Fedspeak with a skill similar to that of a professional analyst. And, in this way, you can interpret that the tone of the monetary policy communication augurs a dovish (expansive monetary policy) or hawkish (restrictive monetary policy) stance.
The researchers used the ChatGPT-3 version and expect the next version to be more refined, but he cautions that it is not foolproof.
“While GPT models may not be able to completely replace human evaluators, they can serve as a valuable tool to assist researchers and analysts in this domain,” they conclude in the paper. article Anne Lundgaard Hansen and Sofia Kazinnik.