Dynamic factor analysis of nonstationary multivariate time series

Peter C.M. Molenaar, Jan G. De Gooijer, Bernhard Schmitz

Research output: Contribution to journalArticle

56 Citations (Scopus)

Abstract

A dynamic factor model is proposed for the analysis of multivariate nonstationary time series in the time domain. The nonstationarity in the series is represented by a linear time dependent mean function. This mild form of nonstationarity is often relevant in analyzing socio-economic time series met in practice. Through the use of an extended version of Molenaar's stationary dynamic factor analysis method, the effect of nonstationarity on the latent factor series is incorporated in the dynamic nonstationary factor model (DNFM). It is shown that the estimation of the unknown parameters in this model can be easily carried out by reformulating the DNFM as a covariance structure model and adopting the ML algorithm proposed by Jöreskog. Furthermore, an empirical example is given to demonstrate the usefulness of the proposed DNFM and the analysis.

Original languageEnglish (US)
Pages (from-to)333-349
Number of pages17
JournalPsychometrika
Volume57
Issue number3
DOIs
StatePublished - Sep 1 1992

Fingerprint

Non-stationary Time Series
Multivariate Time Series
Nonstationarity
Factor Models
Factor analysis
Factor Analysis
Dynamic Analysis
Statistical Factor Analysis
Time series
Dynamic Factor Model
Series
Covariance Structure
Unknown Parameters
Linear Time
Time Domain
Economics
Model structures
Multivariate Analysis
Model
Demonstrate

All Science Journal Classification (ASJC) codes

  • Psychology(all)
  • Applied Mathematics

Cite this

Molenaar, Peter C.M. ; De Gooijer, Jan G. ; Schmitz, Bernhard. / Dynamic factor analysis of nonstationary multivariate time series. In: Psychometrika. 1992 ; Vol. 57, No. 3. pp. 333-349.
@article{85d9960f4c134dc49cd1eff0a6be409a,
title = "Dynamic factor analysis of nonstationary multivariate time series",
abstract = "A dynamic factor model is proposed for the analysis of multivariate nonstationary time series in the time domain. The nonstationarity in the series is represented by a linear time dependent mean function. This mild form of nonstationarity is often relevant in analyzing socio-economic time series met in practice. Through the use of an extended version of Molenaar's stationary dynamic factor analysis method, the effect of nonstationarity on the latent factor series is incorporated in the dynamic nonstationary factor model (DNFM). It is shown that the estimation of the unknown parameters in this model can be easily carried out by reformulating the DNFM as a covariance structure model and adopting the ML algorithm proposed by J{\"o}reskog. Furthermore, an empirical example is given to demonstrate the usefulness of the proposed DNFM and the analysis.",
author = "Molenaar, {Peter C.M.} and {De Gooijer}, {Jan G.} and Bernhard Schmitz",
year = "1992",
month = "9",
day = "1",
doi = "10.1007/BF02295422",
language = "English (US)",
volume = "57",
pages = "333--349",
journal = "Psychometrika",
issn = "0033-3123",
publisher = "Springer New York",
number = "3",

}

Dynamic factor analysis of nonstationary multivariate time series. / Molenaar, Peter C.M.; De Gooijer, Jan G.; Schmitz, Bernhard.

In: Psychometrika, Vol. 57, No. 3, 01.09.1992, p. 333-349.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Dynamic factor analysis of nonstationary multivariate time series

AU - Molenaar, Peter C.M.

AU - De Gooijer, Jan G.

AU - Schmitz, Bernhard

PY - 1992/9/1

Y1 - 1992/9/1

N2 - A dynamic factor model is proposed for the analysis of multivariate nonstationary time series in the time domain. The nonstationarity in the series is represented by a linear time dependent mean function. This mild form of nonstationarity is often relevant in analyzing socio-economic time series met in practice. Through the use of an extended version of Molenaar's stationary dynamic factor analysis method, the effect of nonstationarity on the latent factor series is incorporated in the dynamic nonstationary factor model (DNFM). It is shown that the estimation of the unknown parameters in this model can be easily carried out by reformulating the DNFM as a covariance structure model and adopting the ML algorithm proposed by Jöreskog. Furthermore, an empirical example is given to demonstrate the usefulness of the proposed DNFM and the analysis.

AB - A dynamic factor model is proposed for the analysis of multivariate nonstationary time series in the time domain. The nonstationarity in the series is represented by a linear time dependent mean function. This mild form of nonstationarity is often relevant in analyzing socio-economic time series met in practice. Through the use of an extended version of Molenaar's stationary dynamic factor analysis method, the effect of nonstationarity on the latent factor series is incorporated in the dynamic nonstationary factor model (DNFM). It is shown that the estimation of the unknown parameters in this model can be easily carried out by reformulating the DNFM as a covariance structure model and adopting the ML algorithm proposed by Jöreskog. Furthermore, an empirical example is given to demonstrate the usefulness of the proposed DNFM and the analysis.

UR - http://www.scopus.com/inward/record.url?scp=21144464195&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21144464195&partnerID=8YFLogxK

U2 - 10.1007/BF02295422

DO - 10.1007/BF02295422

M3 - Article

AN - SCOPUS:21144464195

VL - 57

SP - 333

EP - 349

JO - Psychometrika

JF - Psychometrika

SN - 0033-3123

IS - 3

ER -