Working Paper: NBER ID: w2495
Authors: Ray C. Fair
Abstract: This paper presents a way of estimating how accurate VAR models are likely to be for answering structural questions. Data are generated from a dynamic deterministic solution of a structural model; a VAR model is estimated using a subset of these data; and the properties of the VAR model are compared to the properties of the structural model. This procedure has the advantage of eliminating the effects of error terms, since the data are generated from a deterministic simulation. The results show that the VAR models do not seem to be good structural approximations.
Keywords: VAR Models; Structural Approximations; Economic Policy
JEL Codes: No JEL codes provided
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| VAR models do not accurately capture the structural properties of the underlying economic model (C32) | misspecification of VAR models (C32) |
| significant differences in properties of VAR models and structural models (C32) | misspecification of VAR models (C32) |
| VAR models' forecasts contain little information not already captured by the structural model (C53) | inefficiency of VAR models in aggregating information (C32) |
| response properties of VAR models are not reliable for policy analysis (C32) | substantial errors in predicted responses (C52) |
| VAR models significantly underestimate the GNP response (C32) | unreliability of VAR models for policy analysis (C32) |
| VAR models overestimate the money supply responses (E47) | unreliability of VAR models for policy analysis (C32) |