On the Role of the Rank Condition in CCE Estimation of Factor-Augmented Panel Regressions

Joakim Westerlund; Simon Reese; Hande Karabiyik

doi:10.1016/j.jeconom.2016.10.006

On the Role of the Rank Condition in CCE Estimation of Factor-Augmented Panel Regressions

Joakim Westerlund, Simon Reese, Hande Karabiyik

Department of Economics

Research output: Contribution to journal › Article › peer-review

24 Citations (SciVal)

Abstract

A popular approach to factor-augmented panel regressions is the common correlatedeffects (CCE) estimator of Pesaran (Estimation and inference in large heterogeneous panels with a multifactor error structure. Econometrica 74, 967–1012, 2006). This paper points to a problem with the CCE approach that appears in the empirically relevant case when the number of factors is strictly less than the number of observables used in their estimation. Specifically, the use of too many observables causes the second moment matrix of the estimated factors to become asymptotically singular, an issue that has not yet been appropriately accounted for. The purpose of the present paper is to fill this gap in the literature.

Original language	English
Pages (from-to)	60-64
Number of pages	5
Journal	Journal of Econometrics
Volume	197
Issue number	1
Early online date	2016 Oct 28
DOIs	https://doi.org/10.1016/j.jeconom.2016.10.006
Publication status	Published - 2017 Mar

Subject classification (UKÄ)

Economics and Business

Keywords

Factor-augmented panel regression
CCE estimation
Moore–Penrose inverse

Access to Document

10.1016/j.jeconom.2016.10.006

Cite this

@article{4834cd0819ce4f38ae017e118b938417,

title = "On the Role of the Rank Condition in CCE Estimation of Factor-Augmented Panel Regressions",

abstract = "A popular approach to factor-augmented panel regressions is the common correlatedeffects (CCE) estimator of Pesaran (Estimation and inference in large heterogeneous panels with a multifactor error structure. Econometrica 74, 967–1012, 2006). This paper points to a problem with the CCE approach that appears in the empirically relevant case when the number of factors is strictly less than the number of observables used in their estimation. Specifically, the use of too many observables causes the second moment matrix of the estimated factors to become asymptotically singular, an issue that has not yet been appropriately accounted for. The purpose of the present paper is to fill this gap in the literature.",

keywords = "Factor-augmented panel regression, CCE estimation, Moore–Penrose inverse",

author = "Joakim Westerlund and Simon Reese and Hande Karabiyik",

year = "2017",

month = mar,

doi = "10.1016/j.jeconom.2016.10.006",

language = "English",

volume = "197",

pages = "60--64",

journal = "Journal of Econometrics",

issn = "0304-4076",

publisher = "Elsevier",

number = "1",

}

TY - JOUR

T1 - On the Role of the Rank Condition in CCE Estimation of Factor-Augmented Panel Regressions

AU - Westerlund, Joakim

AU - Reese, Simon

AU - Karabiyik, Hande

PY - 2017/3

Y1 - 2017/3

N2 - A popular approach to factor-augmented panel regressions is the common correlatedeffects (CCE) estimator of Pesaran (Estimation and inference in large heterogeneous panels with a multifactor error structure. Econometrica 74, 967–1012, 2006). This paper points to a problem with the CCE approach that appears in the empirically relevant case when the number of factors is strictly less than the number of observables used in their estimation. Specifically, the use of too many observables causes the second moment matrix of the estimated factors to become asymptotically singular, an issue that has not yet been appropriately accounted for. The purpose of the present paper is to fill this gap in the literature.

AB - A popular approach to factor-augmented panel regressions is the common correlatedeffects (CCE) estimator of Pesaran (Estimation and inference in large heterogeneous panels with a multifactor error structure. Econometrica 74, 967–1012, 2006). This paper points to a problem with the CCE approach that appears in the empirically relevant case when the number of factors is strictly less than the number of observables used in their estimation. Specifically, the use of too many observables causes the second moment matrix of the estimated factors to become asymptotically singular, an issue that has not yet been appropriately accounted for. The purpose of the present paper is to fill this gap in the literature.

KW - Factor-augmented panel regression

KW - CCE estimation

KW - Moore–Penrose inverse

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-85007523873&origin=inward&txGid=8323B6AE141346839E1D44651CF90074.wsnAw8kcdt7IPYLO0V48gA%3a90

U2 - 10.1016/j.jeconom.2016.10.006

DO - 10.1016/j.jeconom.2016.10.006

M3 - Article

VL - 197

SP - 60

EP - 64

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 1

ER -

On the Role of the Rank Condition in CCE Estimation of Factor-Augmented Panel Regressions

Abstract

Subject classification (UKÄ)

Keywords

Access to Document

Other files and links

Fingerprint

Estimation and Testing in Panel Data with Cross-Section Dependence

Cite this