Authors: Juan Tenorio, Wilder Perez

Published on: February 06, 2024

Impact Score: 8.38

Arxiv code: Arxiv:2402.04165

Summary

• What is new: The application of ML nowcasting models to GDP growth using a mix of structured macroeconomic and unstructured sentiment data for Peru, demonstrating enhanced prediction accuracy.
• Why this is important: The need for more accurate GDP growth prediction methods in periods of high economic uncertainty.
• What the research proposes: ML-based GDP growth projection models incorporating both structured and high-frequency unstructured sentiment data.
• Results: A 20% to 25% reduction in prediction errors using ML models over traditional methods.

Technical Details

Technological frameworks used: Gradient Boosting Machine, LASSO, Elastic Net

Models used: AR, Dynamic Factor Models

Data used: 91 leading economic indicators from January 2007 to May 2023

Potential Impact

Financial sectors, economic forecasting firms, and policy-making entities could drastically transform or benefit from these insights.

Want to implement this idea in a business?

We have generated a startup concept here: EconNow.