A multilevel factor model: Identification, asymptotic theory and applications

In Choi, Dukpa Kim, Yun Jung Kim, Noh Sun Kwark

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

This paper studies a multilevel factor model with global and country factors. The global factors affect all individuals, whereas the country factors affect only those within each specific country. A sequential procedure to identify the global and country factors separately is proposed. In the initial step, the global factors are estimated by canonical correlation analysis. Using this initial estimator, the principal component estimators (PCEs) of the global and country factors are constructed. It is shown that the PCEs estimate the spaces of the global and country factors consistently and are normally distributed in the limit. Several information criteria that can estimate the number of country factors are proposed. The number of global factors is assumed to be known. Extensive simulation results demonstrate that the sequential procedure and information criteria work well in finite samples. The method of this paper is applied to 25 OECD countries to identify an international business cycle. It is reported that the method extracts a global factor reasonably well.

Original languageEnglish
Pages (from-to)355-377
Number of pages23
JournalJournal of Applied Econometrics
Volume33
Issue number3
DOIs
Publication statusPublished - 2018 Apr 1

Fingerprint

business cycle
OECD
simulation
Asymptotic theory
Factors
Estimator
Information criterion
Principal components

ASJC Scopus subject areas

  • Social Sciences (miscellaneous)
  • Economics and Econometrics

Cite this

A multilevel factor model : Identification, asymptotic theory and applications. / Choi, In; Kim, Dukpa; Kim, Yun Jung; Kwark, Noh Sun.

In: Journal of Applied Econometrics, Vol. 33, No. 3, 01.04.2018, p. 355-377.

Research output: Contribution to journalArticle

Choi, In ; Kim, Dukpa ; Kim, Yun Jung ; Kwark, Noh Sun. / A multilevel factor model : Identification, asymptotic theory and applications. In: Journal of Applied Econometrics. 2018 ; Vol. 33, No. 3. pp. 355-377.
@article{b573f69626a0455dbf68fd9e144f5aa0,
title = "A multilevel factor model: Identification, asymptotic theory and applications",
abstract = "This paper studies a multilevel factor model with global and country factors. The global factors affect all individuals, whereas the country factors affect only those within each specific country. A sequential procedure to identify the global and country factors separately is proposed. In the initial step, the global factors are estimated by canonical correlation analysis. Using this initial estimator, the principal component estimators (PCEs) of the global and country factors are constructed. It is shown that the PCEs estimate the spaces of the global and country factors consistently and are normally distributed in the limit. Several information criteria that can estimate the number of country factors are proposed. The number of global factors is assumed to be known. Extensive simulation results demonstrate that the sequential procedure and information criteria work well in finite samples. The method of this paper is applied to 25 OECD countries to identify an international business cycle. It is reported that the method extracts a global factor reasonably well.",
author = "In Choi and Dukpa Kim and Kim, {Yun Jung} and Kwark, {Noh Sun}",
year = "2018",
month = "4",
day = "1",
doi = "10.1002/jae.2611",
language = "English",
volume = "33",
pages = "355--377",
journal = "Journal of Applied Econometrics",
issn = "0883-7252",
publisher = "John Wiley and Sons Ltd",
number = "3",

}

TY - JOUR

T1 - A multilevel factor model

T2 - Identification, asymptotic theory and applications

AU - Choi, In

AU - Kim, Dukpa

AU - Kim, Yun Jung

AU - Kwark, Noh Sun

PY - 2018/4/1

Y1 - 2018/4/1

N2 - This paper studies a multilevel factor model with global and country factors. The global factors affect all individuals, whereas the country factors affect only those within each specific country. A sequential procedure to identify the global and country factors separately is proposed. In the initial step, the global factors are estimated by canonical correlation analysis. Using this initial estimator, the principal component estimators (PCEs) of the global and country factors are constructed. It is shown that the PCEs estimate the spaces of the global and country factors consistently and are normally distributed in the limit. Several information criteria that can estimate the number of country factors are proposed. The number of global factors is assumed to be known. Extensive simulation results demonstrate that the sequential procedure and information criteria work well in finite samples. The method of this paper is applied to 25 OECD countries to identify an international business cycle. It is reported that the method extracts a global factor reasonably well.

AB - This paper studies a multilevel factor model with global and country factors. The global factors affect all individuals, whereas the country factors affect only those within each specific country. A sequential procedure to identify the global and country factors separately is proposed. In the initial step, the global factors are estimated by canonical correlation analysis. Using this initial estimator, the principal component estimators (PCEs) of the global and country factors are constructed. It is shown that the PCEs estimate the spaces of the global and country factors consistently and are normally distributed in the limit. Several information criteria that can estimate the number of country factors are proposed. The number of global factors is assumed to be known. Extensive simulation results demonstrate that the sequential procedure and information criteria work well in finite samples. The method of this paper is applied to 25 OECD countries to identify an international business cycle. It is reported that the method extracts a global factor reasonably well.

UR - http://www.scopus.com/inward/record.url?scp=85040607275&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85040607275&partnerID=8YFLogxK

U2 - 10.1002/jae.2611

DO - 10.1002/jae.2611

M3 - Article

AN - SCOPUS:85040607275

VL - 33

SP - 355

EP - 377

JO - Journal of Applied Econometrics

JF - Journal of Applied Econometrics

SN - 0883-7252

IS - 3

ER -