Uncertain Prior Economic Knowledge and Statistically Identified Structural Vector Autoregressions
| Autor: | Keweloh, Sascha A. |
|---|---|
| Rok vydání: | 2023 |
| Předmět: | |
| DOI: | 10.48550/arxiv.2303.13281 |
| Popis: | This study proposes an estimator that combines statistical identification with economically motivated restrictions on the interactions. The estimator is identified by (mean) independent non-Gaussian shocks and allows for incorporation of uncertain prior economic knowledge through an adaptive ridge penalty. The estimator shrinks towards economically motivated restrictions when the data is consistent with them and stops shrinkage when the data provides evidence against the restriction. The estimator is applied to analyze the interaction between the stock and oil market. The results suggest that what is usually identified as oil-specific demand shocks can actually be attributed to information shocks extracted from the stock market, which explain about 30-40% of the oil price variation. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|