Testing for a unit root in an arima(p,l,0) signal observed with ma(q) noise

Dong Wan Shin; Sahadeb Sarkar

doi:10.1080/03610929408831408

Testing for a unit root in an arima(p,l,0) signal observed with ma(q) noise

Dong Wan Shin, Sahadeb Sarkar

Department of Statistics

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

2 Scopus citations

Abstract

An ARIMA(p,l,0) signal disturbed by MA(q) noise is an ARIMA(p,l, p+q+1) process restricted by nonlinear constraints on parameters. For this model with a unit root the restricted maximum likelihood estimator (RMLE) of the unit root is strongly consistent and it has the same limiting distribution as the ordinary least squares estimator of the unit root in an AR(1) model tabulated by Dickey and Fuller (1979). A modified RMLE is proposed which has the same limiting properties as the RMLE and is computationally much simpler. Simulation results show that our unit root tests based on the modified RMLE perform very well for small samples and compare favorably with the Said and Dickey (1985) tests with respect to both sizes and powers. An illustrative example from sample survey is given.

Original language	English
Pages (from-to)	2643-2670
Number of pages	28
Journal	Communications in Statistics - Theory and Methods
Volume	23
Issue number	9
DOIs	https://doi.org/10.1080/03610929408831408
State	Published - 1 Jan 1994

Bibliographical note

Funding Information:
The research of the first author was supported by Korea Research Foundation. The research of the second author was partly supported by the Dean's Incentive Grant from the College of Arts and Sciences and a Grant from the College of Arts and Sciences at Oklahoma State University. The authors thank Professor Wayne A. Fuller for kindly suggesting the problem and providing the two examples and the US unemployment dataset, and Dr. Abdoulaye Adam for providing the estimates of the autocovariances of the sampling error. The authors also thank a referee and an associate editor whose comments led to an improved version of this article.

Keywords

Monte Carlo study
large sample
maximum likelihood estimation
measurement error
restricted ARIMA model

Access to Document

10.1080/03610929408831408

Cite this

@article{6fa97ea80a51450e8198008675338aff,

title = "Testing for a unit root in an arima(p,l,0) signal observed with ma(q) noise",

abstract = "An ARIMA(p,l,0) signal disturbed by MA(q) noise is an ARIMA(p,l, p+q+1) process restricted by nonlinear constraints on parameters. For this model with a unit root the restricted maximum likelihood estimator (RMLE) of the unit root is strongly consistent and it has the same limiting distribution as the ordinary least squares estimator of the unit root in an AR(1) model tabulated by Dickey and Fuller (1979). A modified RMLE is proposed which has the same limiting properties as the RMLE and is computationally much simpler. Simulation results show that our unit root tests based on the modified RMLE perform very well for small samples and compare favorably with the Said and Dickey (1985) tests with respect to both sizes and powers. An illustrative example from sample survey is given.",

keywords = "Monte Carlo study, large sample, maximum likelihood estimation, measurement error, restricted ARIMA model",

author = "Shin, {Dong Wan} and Sahadeb Sarkar",

note = "Funding Information: The research of the first author was supported by Korea Research Foundation. The research of the second author was partly supported by the Dean's Incentive Grant from the College of Arts and Sciences and a Grant from the College of Arts and Sciences at Oklahoma State University. The authors thank Professor Wayne A. Fuller for kindly suggesting the problem and providing the two examples and the US unemployment dataset, and Dr. Abdoulaye Adam for providing the estimates of the autocovariances of the sampling error. The authors also thank a referee and an associate editor whose comments led to an improved version of this article.",

year = "1994",

month = jan,

day = "1",

doi = "10.1080/03610929408831408",

language = "English",

volume = "23",

pages = "2643--2670",

journal = "Communications in Statistics - Theory and Methods",

issn = "0361-0926",

publisher = "Taylor and Francis Ltd.",

number = "9",

}

TY - JOUR

T1 - Testing for a unit root in an arima(p,l,0) signal observed with ma(q) noise

AU - Shin, Dong Wan

AU - Sarkar, Sahadeb

N1 - Funding Information: The research of the first author was supported by Korea Research Foundation. The research of the second author was partly supported by the Dean's Incentive Grant from the College of Arts and Sciences and a Grant from the College of Arts and Sciences at Oklahoma State University. The authors thank Professor Wayne A. Fuller for kindly suggesting the problem and providing the two examples and the US unemployment dataset, and Dr. Abdoulaye Adam for providing the estimates of the autocovariances of the sampling error. The authors also thank a referee and an associate editor whose comments led to an improved version of this article.

PY - 1994/1/1

Y1 - 1994/1/1

N2 - An ARIMA(p,l,0) signal disturbed by MA(q) noise is an ARIMA(p,l, p+q+1) process restricted by nonlinear constraints on parameters. For this model with a unit root the restricted maximum likelihood estimator (RMLE) of the unit root is strongly consistent and it has the same limiting distribution as the ordinary least squares estimator of the unit root in an AR(1) model tabulated by Dickey and Fuller (1979). A modified RMLE is proposed which has the same limiting properties as the RMLE and is computationally much simpler. Simulation results show that our unit root tests based on the modified RMLE perform very well for small samples and compare favorably with the Said and Dickey (1985) tests with respect to both sizes and powers. An illustrative example from sample survey is given.

AB - An ARIMA(p,l,0) signal disturbed by MA(q) noise is an ARIMA(p,l, p+q+1) process restricted by nonlinear constraints on parameters. For this model with a unit root the restricted maximum likelihood estimator (RMLE) of the unit root is strongly consistent and it has the same limiting distribution as the ordinary least squares estimator of the unit root in an AR(1) model tabulated by Dickey and Fuller (1979). A modified RMLE is proposed which has the same limiting properties as the RMLE and is computationally much simpler. Simulation results show that our unit root tests based on the modified RMLE perform very well for small samples and compare favorably with the Said and Dickey (1985) tests with respect to both sizes and powers. An illustrative example from sample survey is given.

KW - Monte Carlo study

KW - large sample

KW - maximum likelihood estimation

KW - measurement error

KW - restricted ARIMA model

UR - http://www.scopus.com/inward/record.url?scp=0347433768&partnerID=8YFLogxK

U2 - 10.1080/03610929408831408

DO - 10.1080/03610929408831408

M3 - Article

AN - SCOPUS:0347433768

SN - 0361-0926

VL - 23

SP - 2643

EP - 2670

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

IS - 9

ER -

Testing for a unit root in an arima(p,l,0) signal observed with ma(q) noise

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this