Can a Data-Rich Environment Help Identify the Sources of Model Misspecification?

Abstract

This paper proposes a method for detecting the sources of misspecification in a DSGE model based on testing, in a data-rich environment, the exogeneity of the variables of the DSGE with respect to some auxiliary variables. Finding evidence of non-exogeneity implies misspecification, but finding that some specific variables help predict certain shocks can shed light on the dimensions along which the model is misspecified. Forecast error variance decomposition analysis then helps assess the relevance of the missing channels. The paper puts the proposed methodology to work both in a controlled experiment -by running a Monte Carlo simulations with a known DGP -and using a state-of-the-art model and US data up to 2011. JEL Classification:C32, C52

Dealing with Misspecification in DSGE Models: A Survey

“…In particular, to distinguish possible misspeci…cations in the state-space representation. Originally, Sargent (1989) and Ireland (2004) On one side, Monti (2015) proposes to model the states of the DSGE and auxiliary variables jointly, imposing the restrictions implied by the DSGE as priors, and then verify how much weight is given to the priors in the estimation. Hence, using the Granger-causality test 10 on some auxiliary variables, the researcher can verify if the driving processes of the model are assumed to be exogenous in the DSGE, hence there is some form of misspeci…cation 11 .…”

“…1 1 The approach presented in Monti (2015) is close to the method illustrated by Giannone and Reichlin (2006) to empirically investigate if the shocks recovered from the estimates of a structural VAR are truly structural, which is possible only if the shocks are fundamental. The non-fundalmentalness, as described in Giannone and Reichlin (2006), can be identi…ed by testing whether the VAR is (weakly) exogenous with respect to potentially relevant additional blocks of variables.…”