Consistent estimation of time-varying loadings in high-dimensional factor models

Jakob Guldbæk Mikkelsen; Eric Hillebrand; Giovanni Urga

doi:10.1016/j.jeconom.2018.09.020

Consistent estimation of time-varying loadings in high-dimensional factor models

Jakob Guldbæk Mikkelsen, Eric Hillebrand^*, Giovanni Urga

^*Corresponding author af dette arbejde

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avis › Tidsskriftartikel › Forskning › peer review

13 Citationer (Scopus)

Abstract

In this paper, we develop a two-step maximum likelihood estimator of time-varying loadings in high-dimensional factor models. We specify the loadings to evolve as stationary vector autoregressions (VAR) and show that consistent estimates of the loadings parameters can be obtained. In the first step, principal components are extracted from the data to form factor estimates. In the second step, the parameters of the loadings VARs are estimated as a set of linear regression models with time-varying coefficients. We document the finite-sample properties of the maximum likelihood estimator through an extensive simulation study and illustrate the empirical relevance of the time-varying loadings structure using a large quarterly dataset for the US economy.

Originalsprog	Engelsk
Tidsskrift	Journal of Econometrics
Vol/bind	208
Nummer	2
Sider (fra-til)	535-562
Antal sider	28
ISSN	0304-4076
DOI	https://doi.org/10.1016/j.jeconom.2018.09.020
Status	Udgivet - feb. 2019

Adgang til dokumentet

10.1016/j.jeconom.2018.09.020

Tjek tilgængelighed hos det Kgl. Bibliotek

Andre filer og links

http://www.scopus.com/inward/record.url?scp=85057281290&partnerID=8YFLogxK

Citationsformater

@article{54107fa0f777430e9d03e5e33fc23dc0,

title = "Consistent estimation of time-varying loadings in high-dimensional factor models",

abstract = "In this paper, we develop a two-step maximum likelihood estimator of time-varying loadings in high-dimensional factor models. We specify the loadings to evolve as stationary vector autoregressions (VAR) and show that consistent estimates of the loadings parameters can be obtained. In the first step, principal components are extracted from the data to form factor estimates. In the second step, the parameters of the loadings VARs are estimated as a set of linear regression models with time-varying coefficients. We document the finite-sample properties of the maximum likelihood estimator through an extensive simulation study and illustrate the empirical relevance of the time-varying loadings structure using a large quarterly dataset for the US economy.",

keywords = "Consistent estimation, Factor models, Maximum likelihood estimation, Principal components, Time-varying loadings, Two-step estimation",

author = "Mikkelsen, {Jakob Guldb{\ae}k} and Eric Hillebrand and Giovanni Urga",

note = "AM hentet fra Elsevier",

year = "2019",

month = feb,

doi = "10.1016/j.jeconom.2018.09.020",

language = "English",

volume = "208",

pages = "535--562",

journal = "Journal of Econometrics",

issn = "0304-4076",

publisher = "Elsevier B.V.",

number = "2",

}

TY - JOUR

T1 - Consistent estimation of time-varying loadings in high-dimensional factor models

AU - Mikkelsen, Jakob Guldbæk

AU - Hillebrand, Eric

AU - Urga, Giovanni

N1 - AM hentet fra Elsevier

PY - 2019/2

Y1 - 2019/2

N2 - In this paper, we develop a two-step maximum likelihood estimator of time-varying loadings in high-dimensional factor models. We specify the loadings to evolve as stationary vector autoregressions (VAR) and show that consistent estimates of the loadings parameters can be obtained. In the first step, principal components are extracted from the data to form factor estimates. In the second step, the parameters of the loadings VARs are estimated as a set of linear regression models with time-varying coefficients. We document the finite-sample properties of the maximum likelihood estimator through an extensive simulation study and illustrate the empirical relevance of the time-varying loadings structure using a large quarterly dataset for the US economy.

AB - In this paper, we develop a two-step maximum likelihood estimator of time-varying loadings in high-dimensional factor models. We specify the loadings to evolve as stationary vector autoregressions (VAR) and show that consistent estimates of the loadings parameters can be obtained. In the first step, principal components are extracted from the data to form factor estimates. In the second step, the parameters of the loadings VARs are estimated as a set of linear regression models with time-varying coefficients. We document the finite-sample properties of the maximum likelihood estimator through an extensive simulation study and illustrate the empirical relevance of the time-varying loadings structure using a large quarterly dataset for the US economy.

KW - Consistent estimation

KW - Factor models

KW - Maximum likelihood estimation

KW - Principal components

KW - Time-varying loadings

KW - Two-step estimation

UR - http://www.scopus.com/inward/record.url?scp=85057281290&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2018.09.020

DO - 10.1016/j.jeconom.2018.09.020

M3 - Journal article

AN - SCOPUS:85057281290

SN - 0304-4076

VL - 208

SP - 535

EP - 562

JO - Journal of Econometrics

JF - Journal of Econometrics

IS - 2

ER -

Consistent estimation of time-varying loadings in high-dimensional factor models

Abstract

Adgang til dokumentet

Tjek tilgængelighed hos det Kgl. Bibliotek

Andre filer og links

Fingeraftryk

Citationsformater