Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/333509 
Authors: 
Year of Publication: 
2025
Series/Report no.: 
KOF Working Papers No. 526
Publisher: 
ETH Zurich, KOF Swiss Economic Institute, Zurich
Abstract: 
This article develops a novel economic sentiment indicator (LLM-ESI) by applying large language models to open-ended responses from Swiss business tendency surveys. Using a BERT-based transformer model, it extracts firmlevel sentiment from free-text survey comments and aggregates it into a highfrequency indicator of macroeconomic conditions. The LLM-ESI closely tracks the business cycle and performs on par with, or better than, traditional benchmarks in nowcasting GDP. These results highlight the potential of large language models and open-ended survey responses to deliver timely and nuanced signals for real-time economic analysis.
Subjects: 
Economic Sentiment
Large Language Model
Business Tendency Surveys
Survey Comments
Textual Analysis
Forecasting
JEL: 
C55
C53
E32
E37
E66
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.