Working Paper: CEPR ID: DP7621
Authors: Vo Phuong Mai Le; Patrick Minford; Michael R. Wickens
Abstract: We review the methods used in many papers to evaluate DSGE models by comparing their simulated moments and other features with data equivalents. We note that they select, scale and characterise the shocks without reference to the data; crucially they fail to use the joint distribution of the features under comparison. We illustrate this point by recomputing an assessment of a two-country model in a recent paper; we find that the paper's conclusions are essentially reversed.
Keywords: anomaly; bootstrap; DSGE; indirect inference; puzzle; USEU model; VAR; Wald statistic
JEL Codes: C12; C32; C52; E1
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| limited set of shocks (D80) | flawed evaluation of models (C52) |
| considering all shocks (D81) | improved model performance (C52) |
| joint distribution of model descriptors (C46) | accurate evaluation of model fit (C52) |
| joint behavior of features (C92) | alignment with model predictions (C52) |
| failure to replicate joint correlation (C59) | significant anomaly (C20) |
| model restrictions on joint behavior (D10) | do not reflect real-world data (Y10) |