A conditional-heteroskedasticity-robust confidence interval for the autoregressive parameter

Donald W.K. Andrews, Patrik Guggenberger

Research output: Contribution to journalComment/debate

6 Citations (Scopus)

Abstract

This paper introduces a new confidence interval (CI) for the autoregressive parameter (AR) in an AR(1) model that allows for conditional heteroskedasticity of a general form and AR parameters that are less than or equal to unity. The CI is a modification of Mikusheva's (2007a) modification of Stock's (1991) CI that employs the least squares estimator and a heteroskedasticity-robust variance estimator. The CI is shown to have correct asymptotic size and to be asymptotically similar (in a uniform sense). It does not require any tuning parameters. No existing procedures have these properties. Monte Carlo simulations show that the CI performs well in finite samples in terms of coverage probability and average length, for innovations with and without conditional heteroskedasticity.

Original languageEnglish (US)
Pages (from-to)376-381
Number of pages6
JournalReview of Economics and Statistics
Volume96
Issue number2
DOIs
StatePublished - May 2014

Fingerprint

confidence
coverage
Conditional heteroskedasticity
Confidence interval
innovation
simulation

All Science Journal Classification (ASJC) codes

  • Social Sciences (miscellaneous)
  • Economics and Econometrics

Cite this

@article{5873d598617144a9b4ffa1d2788e698d,
title = "A conditional-heteroskedasticity-robust confidence interval for the autoregressive parameter",
abstract = "This paper introduces a new confidence interval (CI) for the autoregressive parameter (AR) in an AR(1) model that allows for conditional heteroskedasticity of a general form and AR parameters that are less than or equal to unity. The CI is a modification of Mikusheva's (2007a) modification of Stock's (1991) CI that employs the least squares estimator and a heteroskedasticity-robust variance estimator. The CI is shown to have correct asymptotic size and to be asymptotically similar (in a uniform sense). It does not require any tuning parameters. No existing procedures have these properties. Monte Carlo simulations show that the CI performs well in finite samples in terms of coverage probability and average length, for innovations with and without conditional heteroskedasticity.",
author = "Andrews, {Donald W.K.} and Patrik Guggenberger",
year = "2014",
month = "5",
doi = "10.1162/REST_a_00369",
language = "English (US)",
volume = "96",
pages = "376--381",
journal = "Review of Economics and Statistics",
issn = "0034-6535",
publisher = "MIT Press Journals",
number = "2",

}

A conditional-heteroskedasticity-robust confidence interval for the autoregressive parameter. / Andrews, Donald W.K.; Guggenberger, Patrik.

In: Review of Economics and Statistics, Vol. 96, No. 2, 05.2014, p. 376-381.

Research output: Contribution to journalComment/debate

TY - JOUR

T1 - A conditional-heteroskedasticity-robust confidence interval for the autoregressive parameter

AU - Andrews, Donald W.K.

AU - Guggenberger, Patrik

PY - 2014/5

Y1 - 2014/5

N2 - This paper introduces a new confidence interval (CI) for the autoregressive parameter (AR) in an AR(1) model that allows for conditional heteroskedasticity of a general form and AR parameters that are less than or equal to unity. The CI is a modification of Mikusheva's (2007a) modification of Stock's (1991) CI that employs the least squares estimator and a heteroskedasticity-robust variance estimator. The CI is shown to have correct asymptotic size and to be asymptotically similar (in a uniform sense). It does not require any tuning parameters. No existing procedures have these properties. Monte Carlo simulations show that the CI performs well in finite samples in terms of coverage probability and average length, for innovations with and without conditional heteroskedasticity.

AB - This paper introduces a new confidence interval (CI) for the autoregressive parameter (AR) in an AR(1) model that allows for conditional heteroskedasticity of a general form and AR parameters that are less than or equal to unity. The CI is a modification of Mikusheva's (2007a) modification of Stock's (1991) CI that employs the least squares estimator and a heteroskedasticity-robust variance estimator. The CI is shown to have correct asymptotic size and to be asymptotically similar (in a uniform sense). It does not require any tuning parameters. No existing procedures have these properties. Monte Carlo simulations show that the CI performs well in finite samples in terms of coverage probability and average length, for innovations with and without conditional heteroskedasticity.

UR - http://www.scopus.com/inward/record.url?scp=84900835480&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84900835480&partnerID=8YFLogxK

U2 - 10.1162/REST_a_00369

DO - 10.1162/REST_a_00369

M3 - Comment/debate

AN - SCOPUS:84900835480

VL - 96

SP - 376

EP - 381

JO - Review of Economics and Statistics

JF - Review of Economics and Statistics

SN - 0034-6535

IS - 2

ER -