Limit theory for explosive autoregression under conditional heteroskedasticity

Ji Hyung Lee

doi:10.1016/j.jspi.2017.10.008

Limit theory for explosive autoregression under conditional heteroskedasticity

Ji Hyung Lee

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

Abstract

This paper studies an explosive autoregression with conditionally heteroskedastic innovations. The asymptotic distributions of LS, GLS, t-statistics, heteroskedasticity-consistent t-statistics and GLS t-statistics are derived for nonstationary local-to-unity and mildly explosive roots, in which ρ_n satisfies n1−ρ_n→−∞,0. Combined with the existing results, we can cover the range of ρ_n such that n1−ρ_n→−∞,∞. Some results on pure explosive cases are also discussed.

Original language	English (US)
Pages (from-to)	30-55
Number of pages	26
Journal	Journal of Statistical Planning and Inference
Volume	196
DOIs	https://doi.org/10.1016/j.jspi.2017.10.008
State	Published - Aug 2018

Keywords

Asymptotics
Autoregression
Conditional heteroskedasticity
Explosive process
Generalized least squares

ASJC Scopus subject areas

Statistics and Probability
Statistics, Probability and Uncertainty
Applied Mathematics

Online availability

10.1016/j.jspi.2017.10.008

Library availability

Discover UIUC Full Text

Cite this

@article{f44f41e64aa9426b9da9b15852604247,

title = "Limit theory for explosive autoregression under conditional heteroskedasticity",

abstract = "This paper studies an explosive autoregression with conditionally heteroskedastic innovations. The asymptotic distributions of LS, GLS, t-statistics, heteroskedasticity-consistent t-statistics and GLS t-statistics are derived for nonstationary local-to-unity and mildly explosive roots, in which ρn satisfies n1−ρn→−∞,0. Combined with the existing results, we can cover the range of ρn such that n1−ρn→−∞,∞. Some results on pure explosive cases are also discussed.",

keywords = "Asymptotics, Autoregression, Conditional heteroskedasticity, Explosive process, Generalized least squares",

author = "Lee, \{Ji Hyung\}",

note = "Publisher Copyright: {\textcopyright} 2017 Elsevier B.V.",

year = "2018",

month = aug,

doi = "10.1016/j.jspi.2017.10.008",

language = "English (US)",

volume = "196",

pages = "30--55",

journal = "Journal of Statistical Planning and Inference",

issn = "0378-3758",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Limit theory for explosive autoregression under conditional heteroskedasticity

AU - Lee, Ji Hyung

PY - 2018/8

Y1 - 2018/8

N2 - This paper studies an explosive autoregression with conditionally heteroskedastic innovations. The asymptotic distributions of LS, GLS, t-statistics, heteroskedasticity-consistent t-statistics and GLS t-statistics are derived for nonstationary local-to-unity and mildly explosive roots, in which ρn satisfies n1−ρn→−∞,0. Combined with the existing results, we can cover the range of ρn such that n1−ρn→−∞,∞. Some results on pure explosive cases are also discussed.

AB - This paper studies an explosive autoregression with conditionally heteroskedastic innovations. The asymptotic distributions of LS, GLS, t-statistics, heteroskedasticity-consistent t-statistics and GLS t-statistics are derived for nonstationary local-to-unity and mildly explosive roots, in which ρn satisfies n1−ρn→−∞,0. Combined with the existing results, we can cover the range of ρn such that n1−ρn→−∞,∞. Some results on pure explosive cases are also discussed.

KW - Asymptotics

KW - Autoregression

KW - Conditional heteroskedasticity

KW - Explosive process

KW - Generalized least squares

UR - https://www.scopus.com/pages/publications/85034421933

UR - https://www.scopus.com/pages/publications/85034421933#tab=citedBy

U2 - 10.1016/j.jspi.2017.10.008

DO - 10.1016/j.jspi.2017.10.008

M3 - Article

AN - SCOPUS:85034421933

SN - 0378-3758

VL - 196

SP - 30

EP - 55

JO - Journal of Statistical Planning and Inference

JF - Journal of Statistical Planning and Inference

ER -

Limit theory for explosive autoregression under conditional heteroskedasticity

Abstract

Keywords

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this