Staff Reports
Online Estimation of DSGE Models
Number 893
August 2019

JEL classification: C11, C32, C53, E32, E37, E52

Authors: Michael Cai, Marco Del Negro, Edward Herbst, Ethan Matlin, Reca Sarfati, and Frank Schorfheide

This paper illustrates the usefulness of sequential Monte Carlo (SMC) methods in approximating DSGE model posterior distributions. We show how the tempering schedule can be chosen adaptively, explore the benefits of an SMC variant we call generalized tempering for “online” estimation, and provide examples of multimodal posteriors that are well captured by SMC methods. We then use the online estimation of the DSGE model to compute pseudo-out-of-sample density forecasts of DSGE models with and without financial frictions and document the benefits of conditioning DSGE model forecasts on nowcasts of macroeconomic variables and interest rate expectations. We also study whether the predictive ability of DSGE models changes when we use priors that are substantially looser than those commonly adopted in the literature.

Available only in PDF
AUTHOR DISCLOSURE STATEMENT(S)
Michael Cai
The author declares that he has no relevant or material financial interests that relate to the research described in this paper.

Marco Del Negro
The author declares that he has no relevant or material financial interests that relate to the research described in this paper, other than the fact that the author is an employee of the Federal Reserve Bank of New York.

Edward Herbst
The author declares that he has no relevant or material financial interests that relate to the research described in this paper, other than the fact that the author is an employee of the Federal Reserve Board of Governors.

Ethan Matlin
The author declares that he has no relevant or material financial interests that relate to the research described in this paper, other than the fact that the author is an employee of the Federal Reserve Bank of New York.

Rebecca Sarfati
The author declares that she has no relevant or material financial interests that relate to the research described in this paper, other than the fact that the author is an employee of the Federal Reserve Bank of New York.

Frank Schorfheide
The author declares that he received financial support from the National Science Foundation under grant SES 1851634. Beyond that, he has no relevant or material financial interests that relate to the research described in this paper.