New tests for unit roots in autoregressive processes with possibly infinite variance errors

Dong Wan Shin; Beong Soo So

doi:10.1016/S0167-7152(99)00031-0

New tests for unit roots in autoregressive processes with possibly infinite variance errors

Dong Wan Shin, Beong Soo So

Department of Statistics

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

8 Scopus citations

Abstract

For autoregressive processes with possibly infinite variance innovations, tests for unit roots are constructed. The limiting null distributions of the test statistics are standard normal both for finite variance innovations and for infinite variance innovations. The test statistics are the pivotal statistics of modified M-estimators in which the signs of regressors rather than the regressors themselves are used as instrumental variables in estimating unit roots. A Monte-Carlo experiment compares the proposed tests favorably with tests based on the OLSE and tests based on the M-estimators for several innovations.

Original language	English
Pages (from-to)	387-397
Number of pages	11
Journal	Statistics and Probability Letters
Volume	44
Issue number	4
DOIs	https://doi.org/10.1016/S0167-7152(99)00031-0
State	Published - 1 Oct 1999

Bibliographical note

Funding Information:
The authors appreciate the comments of a referee. This research is supported by a grant for women's university from Korea Ministry of Science and Technology.

Keywords

Infinite variance error
M-estimation
Unit root test

Access to Document

10.1016/S0167-7152(99)00031-0

Cite this

@article{b1a6a101c9b24f5a8dabc624f69a8fdf,

title = "New tests for unit roots in autoregressive processes with possibly infinite variance errors",

abstract = "For autoregressive processes with possibly infinite variance innovations, tests for unit roots are constructed. The limiting null distributions of the test statistics are standard normal both for finite variance innovations and for infinite variance innovations. The test statistics are the pivotal statistics of modified M-estimators in which the signs of regressors rather than the regressors themselves are used as instrumental variables in estimating unit roots. A Monte-Carlo experiment compares the proposed tests favorably with tests based on the OLSE and tests based on the M-estimators for several innovations.",

keywords = "Infinite variance error, M-estimation, Unit root test",

author = "Shin, {Dong Wan} and So, {Beong Soo}",

note = "Funding Information: The authors appreciate the comments of a referee. This research is supported by a grant for women's university from Korea Ministry of Science and Technology. ",

year = "1999",

month = oct,

day = "1",

doi = "10.1016/S0167-7152(99)00031-0",

language = "English",

volume = "44",

pages = "387--397",

journal = "Statistics and Probability Letters",

issn = "0167-7152",

publisher = "Elsevier B.V.",

number = "4",

}

TY - JOUR

T1 - New tests for unit roots in autoregressive processes with possibly infinite variance errors

AU - Shin, Dong Wan

AU - So, Beong Soo

N1 - Funding Information: The authors appreciate the comments of a referee. This research is supported by a grant for women's university from Korea Ministry of Science and Technology.

PY - 1999/10/1

Y1 - 1999/10/1

N2 - For autoregressive processes with possibly infinite variance innovations, tests for unit roots are constructed. The limiting null distributions of the test statistics are standard normal both for finite variance innovations and for infinite variance innovations. The test statistics are the pivotal statistics of modified M-estimators in which the signs of regressors rather than the regressors themselves are used as instrumental variables in estimating unit roots. A Monte-Carlo experiment compares the proposed tests favorably with tests based on the OLSE and tests based on the M-estimators for several innovations.

AB - For autoregressive processes with possibly infinite variance innovations, tests for unit roots are constructed. The limiting null distributions of the test statistics are standard normal both for finite variance innovations and for infinite variance innovations. The test statistics are the pivotal statistics of modified M-estimators in which the signs of regressors rather than the regressors themselves are used as instrumental variables in estimating unit roots. A Monte-Carlo experiment compares the proposed tests favorably with tests based on the OLSE and tests based on the M-estimators for several innovations.

KW - Infinite variance error

KW - M-estimation

KW - Unit root test

UR - http://www.scopus.com/inward/record.url?scp=0033211652&partnerID=8YFLogxK

U2 - 10.1016/S0167-7152(99)00031-0

DO - 10.1016/S0167-7152(99)00031-0

M3 - Article

AN - SCOPUS:0033211652

SN - 0167-7152

VL - 44

SP - 387

EP - 397

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

IS - 4

ER -

New tests for unit roots in autoregressive processes with possibly infinite variance errors

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this