Normal tests for unit roots based on instrumental variable estimators

Dong Wan Shin; Beong Soo So

doi:10.1080/02331880310001646635

Normal tests for unit roots based on instrumental variable estimators

Dong Wan Shin, Beong Soo So

Department of Statistics

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

2 Scopus citations

Abstract

For estimating unit roots of autoregressive processes, we introduce a new instrumental variable (IV) method which discounts large values of regressors corresponding to the unit roots. Based on the IV estimator, we propose new unit root tests whose limiting null distributions are standard normal. Observation at time t is adjusted for mean recursively by the sample mean of observations up to the time t. The powers of the proposed tests are better than those of the Dickey-Fuller tests and are comparable to those of the tests based on the weighted symmetric estimator, which are known to have the best power against stationary alternatives.

Original language	English
Pages (from-to)	123-132
Number of pages	10
Journal	Statistics
Volume	38
Issue number	2
DOIs	https://doi.org/10.1080/02331880310001646635
State	Published - Apr 2004

Bibliographical note

Funding Information:
The first author was supported by Korea Research Foundation grant KRF-2002-070-C0001. The second author was supported by KOSEF through the Statistical Research Center for Complex System at Seoul National University.

Keywords

Confidence interval
Instrumental variable estimation
M-estimation
Normal test
Recursive mean adjustment
Unit root test

Access to Document

10.1080/02331880310001646635

Cite this

@article{4e20dbda112d4ad29f4bf88f994c5dd8,

title = "Normal tests for unit roots based on instrumental variable estimators",

abstract = "For estimating unit roots of autoregressive processes, we introduce a new instrumental variable (IV) method which discounts large values of regressors corresponding to the unit roots. Based on the IV estimator, we propose new unit root tests whose limiting null distributions are standard normal. Observation at time t is adjusted for mean recursively by the sample mean of observations up to the time t. The powers of the proposed tests are better than those of the Dickey-Fuller tests and are comparable to those of the tests based on the weighted symmetric estimator, which are known to have the best power against stationary alternatives.",

keywords = "Confidence interval, Instrumental variable estimation, M-estimation, Normal test, Recursive mean adjustment, Unit root test",

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

note = "Funding Information: The first author was supported by Korea Research Foundation grant KRF-2002-070-C0001. The second author was supported by KOSEF through the Statistical Research Center for Complex System at Seoul National University.",

year = "2004",

month = apr,

doi = "10.1080/02331880310001646635",

language = "English",

volume = "38",

pages = "123--132",

journal = "Statistics",

issn = "0233-1888",

publisher = "Taylor and Francis Ltd.",

number = "2",

}

TY - JOUR

T1 - Normal tests for unit roots based on instrumental variable estimators

AU - Shin, Dong Wan

AU - So, Beong Soo

N1 - Funding Information: The first author was supported by Korea Research Foundation grant KRF-2002-070-C0001. The second author was supported by KOSEF through the Statistical Research Center for Complex System at Seoul National University.

PY - 2004/4

Y1 - 2004/4

N2 - For estimating unit roots of autoregressive processes, we introduce a new instrumental variable (IV) method which discounts large values of regressors corresponding to the unit roots. Based on the IV estimator, we propose new unit root tests whose limiting null distributions are standard normal. Observation at time t is adjusted for mean recursively by the sample mean of observations up to the time t. The powers of the proposed tests are better than those of the Dickey-Fuller tests and are comparable to those of the tests based on the weighted symmetric estimator, which are known to have the best power against stationary alternatives.

AB - For estimating unit roots of autoregressive processes, we introduce a new instrumental variable (IV) method which discounts large values of regressors corresponding to the unit roots. Based on the IV estimator, we propose new unit root tests whose limiting null distributions are standard normal. Observation at time t is adjusted for mean recursively by the sample mean of observations up to the time t. The powers of the proposed tests are better than those of the Dickey-Fuller tests and are comparable to those of the tests based on the weighted symmetric estimator, which are known to have the best power against stationary alternatives.

KW - Confidence interval

KW - Instrumental variable estimation

KW - M-estimation

KW - Normal test

KW - Recursive mean adjustment

KW - Unit root test

UR - http://www.scopus.com/inward/record.url?scp=2442611953&partnerID=8YFLogxK

U2 - 10.1080/02331880310001646635

DO - 10.1080/02331880310001646635

M3 - Article

AN - SCOPUS:2442611953

SN - 0233-1888

VL - 38

SP - 123

EP - 132

JO - Statistics

JF - Statistics

IS - 2

ER -

Normal tests for unit roots based on instrumental variable estimators

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this