Surrogate Data Models: Interpreting Large-scale Machine Learning Crisis Prediction Models
Machine learning models are becoming increasingly important in the prediction of economic crises. The models, however, use datasets comprising a large number of predictors (features) which impairs model interpretability and their ability to provide adequate guidance in the design of crisis preventio...
| Main Author: | |
|---|---|
| Other Authors: | , , |
| Format: | eBook |
| Language: | English |
| Published: |
Washington, D.C.
International Monetary Fund
2023
|
| Series: | IMF Working Papers
|
| Subjects: | |
| Online Access: | |
| Collection: | International Monetary Fund - Collection details see MPG.ReNa |
| Summary: | Machine learning models are becoming increasingly important in the prediction of economic crises. The models, however, use datasets comprising a large number of predictors (features) which impairs model interpretability and their ability to provide adequate guidance in the design of crisis prevention and mitigation policies. This paper introduces surrogate data models as dimensionality reduction tools in large-scale crisis prediction models. The appropriateness of this approach is assessed by their application to large-scale crisis prediction models developed at the IMF. The results are consistent with economic intuition and validate the use of surrogates as interpretability tools |
|---|---|
| Physical Description: | 31 pages |
| ISBN: | 9798400234828 |