Markov-Switching Models with Evolving Regime-Specific Parameters: Are Post-War Booms or Recessions All Alike?

45 Pages Posted: 12 Dec 2011 Last revised: 6 Apr 2015

See all articles by Yunjong Eo

Yunjong Eo

Department of Economics, Korea University

Chang-Jin Kim

Dept. of Economics, University of Washington

Date Written: March 5, 2015

Abstract

In this paper, we relax the assumption of constant regime-specific mean growth rates in Hamilton’s (1989) two-state Markov-switching model of the business cycle. We introduce a random walk hierarchical prior for each regime-specific mean growth rate and impose a cointegrating relationship between the mean growth rates in recessionary and expansionary periods. By applying the proposed model to postwar U.S. real GDP growth (1947:Q4-2011:Q3), we uncover the evolving nature of the regime-specific mean growth rates of real output in the U.S. business cycle. Additional features of the postwar U.S. business cycle that we uncover include: i) a steady decline in the long-run mean growth rate of real output over the postwar sample and ii) an asymmetric error-correction mechanism when the economy deviates from its long-run equilibrium.

Keywords: Business Cycle, Evolving Regime-Specific Parameters, Hierarchical Prior, Markov Switching, Error-Correction Dynamics, MCMC, State-Space Model

JEL Classification: C11, C22, C51, E32

Suggested Citation

Eo, Yunjong and Kim, Chang-Jin, Markov-Switching Models with Evolving Regime-Specific Parameters: Are Post-War Booms or Recessions All Alike? (March 5, 2015). Available at SSRN: https://ssrn.com/abstract=1971169 or http://dx.doi.org/10.2139/ssrn.1971169

Yunjong Eo

Department of Economics, Korea University ( email )

1 Anam-dong 5 ka
Seoul, 136-701
Korea, Republic of (South Korea)

Chang-Jin Kim (Contact Author)

Dept. of Economics, University of Washington ( email )

Department of Economics (Box 353330)
University of Washington
Seattle, WA 98195-3330
United States

HOME PAGE: http://https://econ.washington.edu/people/chang-jin-kim

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
159
Abstract Views
952
Rank
306,672
PlumX Metrics