A Forecasting Metric for Evaluating DSGE Models for Policy Analysis
نویسنده
چکیده
This paper applies a new Bayesian framework laid out in Faust and Gupta (2009) for evaluating the suitability of dynamic stochastic general equilibrium (DSGE) models for the task of monetary policy analysis. The paper characterizes practical monetary policy analysis as determining how intended policy should be revised in light of the structural interpretation of incoming news. The news is defined as the one-step ahead forecast errors and the first and second moments of this news sufficiently summarize its structure for the purposes of monetary policy analysis. To shed light on the structural implications of this news, the paper estimates the implied structural shocks from fitting the DSGE model of Smets and Wouters (2007) to data. These fitted shocks then provide a structural interpretation for the observed news. The paper then evaluates the first and second moments of both the forecast errors and structural shocks, finding strengths of the model and important shortcomings. The paper finds that the posterior distribution for the realized value of the mean and cross-correlations of the structural shocks are typically nonzero. This result is surprising because structural shocks by construction are supposed to be zero-mean and uncorrelated. The results, therefore, identify specific frictions and structural aspects of the model that cause misspecification and suggest areas of improvement for model building. JEL Classification: C11, C52, E17, E5
منابع مشابه
How Good Is What You’ve Got? DGSE-VAR as a Toolkit for Evaluating DSGE Models
E C O N O M I C R E V I E W Second Quarter 2006 Dynamic stochastic general equilibrium (DSGE) models are becoming increasingly popular in central banking circles. The number of central bank–sponsored conferences on DSGE modeling and the amount of staff resources devoted to DSGE model development and estimation have risen dramatically over the past five years. This trend has affected monetary po...
متن کاملBayesian forecasting with small and medium scale factor-augmented vector autoregressive DSGE models
Advanced Bayesian methods are employed in estimating dynamic stochastic general equilibrium (DSGE) models. Although policymakers and practitioners are particularly interested in DSGE models, these are typically too stylized to be taken directly to the data and often yield weak prediction results. Hybrid models can deal with some of the DSGE model misspeci cations. Major advances in Bayesian est...
متن کاملOn the predictability of time-varying VAR and DSGE models
Over the last few years, there has been a growing interest in DSGE modelling for predicting macroeconomic uctuations and conducting quantitative policy analysis. Hybrid DSGE models have become popular for dealing with some of the DSGE misspeci cations as they are able to solve the tradeo¤ between theoretical coherence and empirical t. However, these models are still linear and they do not con...
متن کاملEvaluating DSGE Models for Monetary and Fiscal Policy Analysis
This paper evaluates the suitability of current DSGE models for monetary policy analysis from a Bayesian perspective, recognizing that these models may be useful even if they are misspecified along some dimensions, using the framework laid out in Faust and Gupta (2009). The paper argues that practical monetary policy analysis deals with determining how intended policy should be revised in light...
متن کاملModel Validation and DSGE Modeling
The primary objective of this paper is to revisit DSGE models with a view to appraise their statistical adequacy and propose ways to ameliorate their empirical pertinence. This paper brings out several weaknesses of the traditional DSGE modeling, including statistical misspecification, non-identification of deep parameters, substantive inadequacy, weak forecasting performance and potentially mi...
متن کامل