NWO-Open Competition-M grant: The best of both worlds: model specification and inference in hybrid machine learning econometric models

New techniques like deep learning and large language models (LLMs) are increasingly applied in everyday life and policy contexts. Quantifying the uncertainty surrounding predictions based on these models remains an underexplored field. This project pushes the frontier on three fronts by:
1. constructing hybrid models exploiting the best side of econometric and machine learning models and applying them to study the spread of fine particulate matter.
2. constructing bands of uncertainty for the above models;
3. constructing frames of uncertainty for models for unstructured data like text and picture predictions and studying the quality of bank's climate risk reporting.