What you match does matter: the effects of data on DSGE estimation

Abstract

This paper explores the effects of using alternative combinations of observables for the estimation of Dynamic Stochastic General Equilibrium (DSGE) models. I find that the estimation of structural parameters describing the Taylor rule and sticky contracts in prices and wages is particularly sensitive to the set of observables. In terms of the model's predictions, the exclusion of some observables may lead to estimated parameters with unexpected outcomes, such as recessions following a positive technology shock. More importantly, two ways to assess different sets of observables are proposed. These measures favor a dataset consisting of seven observables. Copyright © 2009 John Wiley & Sons, Ltd.

Methods versus substance: Measuring the effects of technology shocks

“…Guerron-Quintana (2010) documents how parameter estimates of a variant of the SmetsWouters model are affected by the choice of observables used in the Bayesian estimation. Iskrev (2010) and Komunjer and Ng (2011) develop formal conditions for the identifiability of DSGE model parameters based on first and second moments of model variables.…”

“…Guerron-Quintana (2010) examines how data choice affects parameter identifiability in the Smets-Wouters model by eliminating subsets of the seven macroeconomic time series that are typically used to estimate the Smets-Wouters model.…”

Financial frictions, trends, and the great recession

“…The empirical treatment used in our paper relates to the extensive literature on the estimation of dynamic stochastic general equilibrium models (Fernandez‐Villaverde, Guerron‐Quintana, and Rubio‐Ramirez () and Guerron‐Quintana ()). We are close to Ajello (), who estimates a model with financial frictions in which competitive financial intermediaries transfer resources between entrepreneurs with heterogeneous skills.…”

On the Sources of Aggregate Fluctuations in Emerging Economies

“…It should be noted that the choice of which variables to use in the estimation and the use or not of measurement errors is indeed not a trivial one. Guerron‐Quintana () has shown that, in the estimation of DSGE models by Bayesian methods, posterior distributions may significantly vary across various sets of observables used. In the working paper version of this work, Chang and Fernandez (), we assess the robustness of our results to cases where no measurement errors are used.…”