2

How much Structure in

Empirical Models?

Fabio Canova

Abstract

This chapter highlights the problems that structural methods and SVAR approaches have when
estimating DSGE models and examining their ability to capture important features of the data.
We show that structural methods are subject to severe identification problems due, in large
part, to the nature of DSGE models. The problems can be patched up in a number of ways,
but solved only if DSGEs are completely reparameterized or respecified. The potential mis-
specification of the structural relationships gives Bayesian methods an edge over classical ones
in structural estimation. SVAR approaches may face invertibility problems but simple diagnos-
tics can help to detect and remedy these problems. A pragmatic empirical approach ought
to use the flexibility of SVARs against potential misspecification of the structural relationships
but must firmly tie SVARs to the class of DSGE models which could have generated the
data.

2.1 Introduction 68
2.2 DSGE models 71
2.2.1 Identification 73
2.2.1.1 Example 1: observational equivalence 75
2.2.1.2 Example 2: identification problems in a New Keynesian
model 76
2.3 Structural VARs 85
2.3.1 Invertibility 88
2.3.1.1 Example 3: a Blanchard and Quah economy 91
2.3.1.2 Example 4: an RBC model 91
2.4 Some final thoughts 93

2.1 Introduction

The 1990s witnessed a remarkable development in the specification of stochastic
general equilibrium models. The literature has added considerable realism to the
popular workhorses of the 1980s; a number of shocks and frictions have been intro-
duced into first-generation Real Business Cycle (RBC) models driven by a single

68