On the asymptotic distribution of the quadratic GMM estimator of a dynamic panel data model under a unit root

Tue Gørgens; Chirok Han; Sen Xue

doi:10.1016/j.econlet.2020.109605

On the asymptotic distribution of the quadratic GMM estimator of a dynamic panel data model under a unit root

Tue Gørgens, Chirok Han, Sen Xue

Department of Economics

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

Abstract

This paper considers the GMM estimator, αˆ, of the autoregressive parameter in linear dynamic panel data models with fixed effects when the data-generating process has a unit root. Previous literature has established that the limit distribution of n^1∕4(αˆ−1) is degenerate and nondegenerate each with probability 1/2. We sharpen this result by showing that the limit distribution of n^1∕2(αˆ−1) is nondegenerate when n^1∕4(αˆ−1) converges in probability to 0, and we characterize the limit distribution which is nonstandard.

Original language	English
Article number	109605
Journal	Economics Letters
Volume	197
DOIs	https://doi.org/10.1016/j.econlet.2020.109605
Publication status	Published - 2020 Dec

Keywords

Dynamic panel data models
Fixed effects
Generalized method of moments
Nonstandard limiting distributions
Quadratic moment restrictions

ASJC Scopus subject areas

Finance
Economics and Econometrics

Access to Document

10.1016/j.econlet.2020.109605

Cite this

@article{01c317b07dad4f46b5ad69ac1fe0805f,

title = "On the asymptotic distribution of the quadratic GMM estimator of a dynamic panel data model under a unit root",

abstract = "This paper considers the GMM estimator, αˆ, of the autoregressive parameter in linear dynamic panel data models with fixed effects when the data-generating process has a unit root. Previous literature has established that the limit distribution of n1∕4(αˆ−1) is degenerate and nondegenerate each with probability 1/2. We sharpen this result by showing that the limit distribution of n1∕2(αˆ−1) is nondegenerate when n1∕4(αˆ−1) converges in probability to 0, and we characterize the limit distribution which is nonstandard.",

keywords = "Dynamic panel data models, Fixed effects, Generalized method of moments, Nonstandard limiting distributions, Quadratic moment restrictions",

author = "Tue G{\o}rgens and Chirok Han and Sen Xue",

note = "Funding Information: This research was previously circulated in a paper entitled “Asymptotic Distributions of the Quadratic GMM Estimator in Linear Dynamic Panel Data Models”. Research by Han was supported by Korea University ( K2016591 ). Publisher Copyright: {\textcopyright} 2020",

year = "2020",

month = dec,

doi = "10.1016/j.econlet.2020.109605",

language = "English",

volume = "197",

journal = "Economics Letters",

issn = "0165-1765",

publisher = "Elsevier",

}

TY - JOUR

T1 - On the asymptotic distribution of the quadratic GMM estimator of a dynamic panel data model under a unit root

AU - Gørgens, Tue

AU - Han, Chirok

AU - Xue, Sen

N1 - Funding Information: This research was previously circulated in a paper entitled “Asymptotic Distributions of the Quadratic GMM Estimator in Linear Dynamic Panel Data Models”. Research by Han was supported by Korea University ( K2016591 ). Publisher Copyright: © 2020

PY - 2020/12

Y1 - 2020/12

N2 - This paper considers the GMM estimator, αˆ, of the autoregressive parameter in linear dynamic panel data models with fixed effects when the data-generating process has a unit root. Previous literature has established that the limit distribution of n1∕4(αˆ−1) is degenerate and nondegenerate each with probability 1/2. We sharpen this result by showing that the limit distribution of n1∕2(αˆ−1) is nondegenerate when n1∕4(αˆ−1) converges in probability to 0, and we characterize the limit distribution which is nonstandard.

AB - This paper considers the GMM estimator, αˆ, of the autoregressive parameter in linear dynamic panel data models with fixed effects when the data-generating process has a unit root. Previous literature has established that the limit distribution of n1∕4(αˆ−1) is degenerate and nondegenerate each with probability 1/2. We sharpen this result by showing that the limit distribution of n1∕2(αˆ−1) is nondegenerate when n1∕4(αˆ−1) converges in probability to 0, and we characterize the limit distribution which is nonstandard.

KW - Dynamic panel data models

KW - Fixed effects

KW - Generalized method of moments

KW - Nonstandard limiting distributions

KW - Quadratic moment restrictions

UR - http://www.scopus.com/inward/record.url?scp=85092499798&partnerID=8YFLogxK

U2 - 10.1016/j.econlet.2020.109605

DO - 10.1016/j.econlet.2020.109605

M3 - Article

AN - SCOPUS:85092499798

SN - 0165-1765

VL - 197

JO - Economics Letters

JF - Economics Letters

M1 - 109605

ER -

On the asymptotic distribution of the quadratic GMM estimator of a dynamic panel data model under a unit root

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this