In 2007 all ECB publications feature a motif taken from the €20 banknote. This paper can be downloaded without charge from

ABSTRACT. A multivariate model, identifying monetary policy and allowing for simultaneity and regime switching in coefficients and variances, is confronted with US data since 1959. The best fit is with a model that allows time variation in structural disturbance variances only. Among models that allow for changes in equation coefficients also, the best fit is for a model that allows coefficients to change only in the monetary policy rule. That model allows switching among three main regimes and one rarely and briefly occurring regime. The three main regimes correspond roughly to periods when most observers believe that monetary policy actually differed, and the differences in policy behavior are substantively interesting, though statistically ill-determined. The estimates imply monetary targeting was central in the early 80’s, but also important sporadically in the 70’s. The changes in regime were essential neither to the rise in inflation in the 70’s nor to its decline in the 80’s. I. THE DEBATE OVER MONETARY POLICY CHANGE In an influential paper, Clarida, Galí and Gertler 2000 (CGG) presented evidence that US monetary policy changed between the 1970’s and the 1980’s, indeed that in the 70’s

This paper considers a prototypical monetary business cycle model for the U.S. economy, in which the equilibrium is undetermined if monetary policy is `passive'. In previous multivariate studies it has been common practice to restrict parameter estimates to values for which the equilibrium is unique. We show how the likelihood-based estimation of dynamic stochastic general equilibrium models can be extended to allow for indeterminacies and sunspot fluctuations. We construct

Abstract: This paper uses a simple New Keynesian monetary DSGE model as a prior for a vector autoregression and shows that the resulting model is competitive with standard benchmarks in terms of forecasting and can be used for policy analysis. JEL classification: C11, C32, C53

This paper studies the properties of the Bayesian approach to estimation and comparison of dynamic equilibrium economies. Both tasks can be performed even if the models are nonnested, misspecified, and nonlinear. First, we show that Bayesian methods have a classical interpretation: asymptotically, the parameter point estimates converge to their pseudotrue values, and the best model under the KullbackLeibler distance will have the highest posterior probability. Second, we illustrate the strong small sample behavior of the approach using a well-known application: the U.S. cattle cycle. Bayesian estimates outperform maximum likelihood results, and the proposed model is easily compared with a set of BVARs.

Schorfheide was visiting the Federal Reserve Bank of Philadelphia, for whose hospitality is thankful. Financial sup-

Abstract not found

Abstract not found

This paper reviews Bayesian methods that have been developed in recent years to estimate and evaluate dynamic stochastic general equilibrium (DSGE) models. We consider the estimation of linearized DSGE models, the evaluation of models based on Bayesian model checking, posterior odds comparisons, and comparisons to vector autoregressions, as well as the nonlinear estimation based on a second-order accurate model solution. These methods are applied to data generated from correctly specified and misspecified linearized DSGE models, and a DSGE model that was solved with a second-order perturbation method. (JEL C11, C32, C51, C52)

This paper compares two methods for undertaking likelihood-based inference in dynamic equilibrium economies: a sequential Monte Carlo filter and the Kalman filter. The sequential Monte Carlo filter exploits the nonlinear structure of the economy and evaluates the likelihood function of the model by simulation methods. The Kalman filter estimates a linearization of the economy around the steady state. We report two main results. First, both for simulated and for real data, the sequential Monte Carlo filter delivers a substantially better fit of the model to the data as measured by the marginal likelihood. This is true even for a nearly linear case. Second, the differences in terms of point estimates, although relatively small in absolute values, have important effects on the moments of the model. We conclude that the nonlinear filter is a superior procedure for taking models to the data. Copyright © 2005 John Wiley & Sons, Ltd. 1.