Macroeconomic and Financial Forecasting in the Aftermath of the Covid Shock: Inflation Forecasting With Large Language Models

Key Points

• The aim is to evaluate the forecasting ability of large language models compared to traditional models in predicting inflation.
• Compared inflation forecasts from large language models against traditional models and Kalman filter approach.
• Utilized monthly inflation data from the United States and euro area.
• Forecast period extended from January 2020 to September 2025.
• All large language models demonstrated a significant improvement in forecasting performance compared to traditional models.
• LLMs provided more accurate inflation forecasts during the study period.