Empirical likelihood test for causality of bivariate AR(1) processes

D. Li; N. H. Chan; L. Peng

doi:10.1017/S0266466613000339

Empirical likelihood test for causality of bivariate AR(1) processes

D. Li
, N. H. Chan
, L. Peng

Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review

7 Citations (Scopus)

Abstract

Testing for causality is of critical importance for many econometric applications. For bivariate AR(1) processes, the limit distributions of causality tests based on least squares estimation depend on the presence of nonstationary processes. When nonstationary processes are present, the limit distributions of such tests are usually very complicated, and the full-sample bootstrap method becomes inconsistent as pointed out in Choi (2005, Statistics and Probability Letters 75, 39-48). In this paper, a profile empirical likelihood method is proposed to test for causality. The proposed test statistic is robust against the presence of nonstationary processes in the sense that one does not have to determine the existence of nonstationary processes a priori. Simulation studies confirm that the proposed test statistic works well.

Original language	English
Pages (from-to)	357-371
Journal	Econometric Theory
Volume	30
Issue number	2
DOIs	https://doi.org/10.1017/S0266466613000339
Publication status	Published - 29 Nov 2014
Externally published	Yes

Bibliographical note

Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].

Access Output

10.1017/S0266466613000339

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@article{b18329cc1f3f4f63819ef04174a3eccd,

title = "Empirical likelihood test for causality of bivariate AR(1) processes",

abstract = "Testing for causality is of critical importance for many econometric applications. For bivariate AR(1) processes, the limit distributions of causality tests based on least squares estimation depend on the presence of nonstationary processes. When nonstationary processes are present, the limit distributions of such tests are usually very complicated, and the full-sample bootstrap method becomes inconsistent as pointed out in Choi (2005, Statistics and Probability Letters 75, 39-48). In this paper, a profile empirical likelihood method is proposed to test for causality. The proposed test statistic is robust against the presence of nonstationary processes in the sense that one does not have to determine the existence of nonstationary processes a priori. Simulation studies confirm that the proposed test statistic works well.",

author = "D. Li and Chan, \{N. H.\} and L. Peng",

note = "Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected]. ",

year = "2014",

month = nov,

day = "29",

doi = "10.1017/S0266466613000339",

language = "English",

volume = "30",

pages = "357--371",

journal = "Econometric Theory",

issn = "0266-4666",

publisher = "Cambridge University Press",

number = "2",

}

TY - JOUR

T1 - Empirical likelihood test for causality of bivariate AR(1) processes

AU - Li, D.

AU - Chan, N. H.

AU - Peng, L.

N1 - Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].

PY - 2014/11/29

Y1 - 2014/11/29

N2 - Testing for causality is of critical importance for many econometric applications. For bivariate AR(1) processes, the limit distributions of causality tests based on least squares estimation depend on the presence of nonstationary processes. When nonstationary processes are present, the limit distributions of such tests are usually very complicated, and the full-sample bootstrap method becomes inconsistent as pointed out in Choi (2005, Statistics and Probability Letters 75, 39-48). In this paper, a profile empirical likelihood method is proposed to test for causality. The proposed test statistic is robust against the presence of nonstationary processes in the sense that one does not have to determine the existence of nonstationary processes a priori. Simulation studies confirm that the proposed test statistic works well.

AB - Testing for causality is of critical importance for many econometric applications. For bivariate AR(1) processes, the limit distributions of causality tests based on least squares estimation depend on the presence of nonstationary processes. When nonstationary processes are present, the limit distributions of such tests are usually very complicated, and the full-sample bootstrap method becomes inconsistent as pointed out in Choi (2005, Statistics and Probability Letters 75, 39-48). In this paper, a profile empirical likelihood method is proposed to test for causality. The proposed test statistic is robust against the presence of nonstationary processes in the sense that one does not have to determine the existence of nonstationary processes a priori. Simulation studies confirm that the proposed test statistic works well.

UR - https://www.scopus.com/pages/publications/84927561724

UR - https://www.scopus.com/pages/publications/84927561724#tab=impact

U2 - 10.1017/S0266466613000339

DO - 10.1017/S0266466613000339

M3 - RGC 21 - Publication in refereed journal

SN - 0266-4666

VL - 30

SP - 357

EP - 371

JO - Econometric Theory

JF - Econometric Theory

IS - 2

ER -

Empirical likelihood test for causality of bivariate AR(1) processes

Abstract

Bibliographical note

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this