Choices between OLS with robust inference and feasible GLS in time series regressions

Richard T. Baillie, Kunho Kim

Research output: Contribution to journalArticle

Abstract

We consider the practice of estimating static regressions by OLS from time series data and using robust standard errors for inference. Depending on the form of exogeneity being violated, the asymptotic bias of OLS can exceed that of GLS. Feasible GLS, where the error process is approximated by a sieve autoregression, can dominate the OLS approach with robust standard errors both in terms of bias and MSE for some regions of the parameter space.

Original languageEnglish
Pages (from-to)218-221
Number of pages4
JournalEconomics Letters
Volume171
DOIs
StatePublished - 2018 Oct 1

Fingerprint

Robust inference
Standard error
Exogeneity
Time series data
Autoregression
Inference
Asymptotic bias

Keywords

  • Asymptotic bias
  • Feasible GLS
  • GLS
  • OLS
  • Robust inference

Cite this

@article{ffb01aae3abe46849d118ab20a9739dc,
title = "Choices between OLS with robust inference and feasible GLS in time series regressions",
abstract = "We consider the practice of estimating static regressions by OLS from time series data and using robust standard errors for inference. Depending on the form of exogeneity being violated, the asymptotic bias of OLS can exceed that of GLS. Feasible GLS, where the error process is approximated by a sieve autoregression, can dominate the OLS approach with robust standard errors both in terms of bias and MSE for some regions of the parameter space.",
keywords = "Asymptotic bias, Feasible GLS, GLS, OLS, Robust inference",
author = "Baillie, {Richard T.} and Kunho Kim",
year = "2018",
month = "10",
day = "1",
doi = "10.1016/j.econlet.2018.07.036",
language = "English",
volume = "171",
pages = "218--221",
journal = "Economics Letters",
issn = "0165-1765",

}

Choices between OLS with robust inference and feasible GLS in time series regressions. / Baillie, Richard T.; Kim, Kunho.

In: Economics Letters, Vol. 171, 01.10.2018, p. 218-221.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Choices between OLS with robust inference and feasible GLS in time series regressions

AU - Baillie, Richard T.

AU - Kim, Kunho

PY - 2018/10/1

Y1 - 2018/10/1

N2 - We consider the practice of estimating static regressions by OLS from time series data and using robust standard errors for inference. Depending on the form of exogeneity being violated, the asymptotic bias of OLS can exceed that of GLS. Feasible GLS, where the error process is approximated by a sieve autoregression, can dominate the OLS approach with robust standard errors both in terms of bias and MSE for some regions of the parameter space.

AB - We consider the practice of estimating static regressions by OLS from time series data and using robust standard errors for inference. Depending on the form of exogeneity being violated, the asymptotic bias of OLS can exceed that of GLS. Feasible GLS, where the error process is approximated by a sieve autoregression, can dominate the OLS approach with robust standard errors both in terms of bias and MSE for some regions of the parameter space.

KW - Asymptotic bias

KW - Feasible GLS

KW - GLS

KW - OLS

KW - Robust inference

UR - http://www.scopus.com/inward/record.url?scp=85051380596&partnerID=8YFLogxK

U2 - 10.1016/j.econlet.2018.07.036

DO - 10.1016/j.econlet.2018.07.036

M3 - Article

AN - SCOPUS:85051380596

VL - 171

SP - 218

EP - 221

JO - Economics Letters

JF - Economics Letters

SN - 0165-1765

ER -