New index of coincident indicators: A multivariate Markov switching factor model approach

Myung-Jig Kim, Ji Sung Yoo

Research output: Contribution to journalArticle

63 Citations (Scopus)

Abstract

This paper extends the univariate Markov switching unobserved component model to the multivariate Markov switching factor model of coincident economic indicators. An approximate ML method is developed to estimate the model. The extracted Markov switching factor may be interpreted as the coincident index and is comparable to the Stock-Watson index which differs only by the Markov switching component. Using four constituent series of the DOC coincident index for the period January 1960 to June 1992, the proposed model generates recessionary and expansionary periods which are remarkably consistent with the NBER chronology of the business cycles from the noisy monthly data. This paper also finds that duration dependence does matter.

Original languageEnglish
Pages (from-to)607-630
Number of pages24
JournalJournal of Monetary Economics
Volume36
Issue number3
DOIs
StatePublished - 1995 Jan 1

Fingerprint

Markov switching
Coincident indicators
Unobserved components model
Economic indicators
Factors
Duration dependence
Business cycles
Chronology
Stock index

Keywords

  • Factor model
  • Kalman filter
  • Markov switching
  • Stock-Watson model
  • Time-varying transition probability

Cite this

@article{54503b3c68194991b7ad374ecc906556,
title = "New index of coincident indicators: A multivariate Markov switching factor model approach",
abstract = "This paper extends the univariate Markov switching unobserved component model to the multivariate Markov switching factor model of coincident economic indicators. An approximate ML method is developed to estimate the model. The extracted Markov switching factor may be interpreted as the coincident index and is comparable to the Stock-Watson index which differs only by the Markov switching component. Using four constituent series of the DOC coincident index for the period January 1960 to June 1992, the proposed model generates recessionary and expansionary periods which are remarkably consistent with the NBER chronology of the business cycles from the noisy monthly data. This paper also finds that duration dependence does matter.",
keywords = "Factor model, Kalman filter, Markov switching, Stock-Watson model, Time-varying transition probability",
author = "Myung-Jig Kim and Yoo, {Ji Sung}",
year = "1995",
month = "1",
day = "1",
doi = "10.1016/0304-3932(95)01229-X",
language = "English",
volume = "36",
pages = "607--630",
journal = "Journal of Monetary Economics",
issn = "0304-3932",
number = "3",

}

New index of coincident indicators : A multivariate Markov switching factor model approach. / Kim, Myung-Jig; Yoo, Ji Sung.

In: Journal of Monetary Economics, Vol. 36, No. 3, 01.01.1995, p. 607-630.

Research output: Contribution to journalArticle

TY - JOUR

T1 - New index of coincident indicators

T2 - A multivariate Markov switching factor model approach

AU - Kim, Myung-Jig

AU - Yoo, Ji Sung

PY - 1995/1/1

Y1 - 1995/1/1

N2 - This paper extends the univariate Markov switching unobserved component model to the multivariate Markov switching factor model of coincident economic indicators. An approximate ML method is developed to estimate the model. The extracted Markov switching factor may be interpreted as the coincident index and is comparable to the Stock-Watson index which differs only by the Markov switching component. Using four constituent series of the DOC coincident index for the period January 1960 to June 1992, the proposed model generates recessionary and expansionary periods which are remarkably consistent with the NBER chronology of the business cycles from the noisy monthly data. This paper also finds that duration dependence does matter.

AB - This paper extends the univariate Markov switching unobserved component model to the multivariate Markov switching factor model of coincident economic indicators. An approximate ML method is developed to estimate the model. The extracted Markov switching factor may be interpreted as the coincident index and is comparable to the Stock-Watson index which differs only by the Markov switching component. Using four constituent series of the DOC coincident index for the period January 1960 to June 1992, the proposed model generates recessionary and expansionary periods which are remarkably consistent with the NBER chronology of the business cycles from the noisy monthly data. This paper also finds that duration dependence does matter.

KW - Factor model

KW - Kalman filter

KW - Markov switching

KW - Stock-Watson model

KW - Time-varying transition probability

UR - http://www.scopus.com/inward/record.url?scp=0000407228&partnerID=8YFLogxK

U2 - 10.1016/0304-3932(95)01229-X

DO - 10.1016/0304-3932(95)01229-X

M3 - Article

AN - SCOPUS:0000407228

VL - 36

SP - 607

EP - 630

JO - Journal of Monetary Economics

JF - Journal of Monetary Economics

SN - 0304-3932

IS - 3

ER -