Model selection using J-test for the spatial autoregressive model vs. the matrix exponential spatial model

Xiaoyi Han; Lung fei Lee

doi:10.1016/j.regsciurbeco.2012.07.005

Model selection using J-test for the spatial autoregressive model vs. the matrix exponential spatial model

Xiaoyi Han^*, Lung fei Lee

^*Corresponding author for this work

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

25 Citations (Scopus)

Abstract

We consider using the J-test procedure for the non-nested model selection problem between the spatial autoregressive (SAR) model and the matrix exponential spatial specification (MESS) model. The 2SLS and GMM methods are used to implement the J-test procedure and derive several test statistics under the GMM framework. We investigate the behavior of those J-test statistics in terms of pseudo true values. We extend the J-test procedure into the setting when error terms in the model are with unknown heteroskedasticity. Monte Carlo results suggest with strong spatial dependence the J-test statistics can have good power to distinguish the SAR and MESS models.

Original language	English
Pages (from-to)	250-271
Number of pages	22
Journal	Regional Science and Urban Economics
Volume	43
Issue number	2
DOIs	https://doi.org/10.1016/j.regsciurbeco.2012.07.005
Publication status	Published - 1 Mar 2013
Externally published	Yes

Bibliographical note

Publisher Copyright:
© 2012 Elsevier B.V.

Keywords

GMM
J-test
Matrix exponential spatial model
Pseudo true value
Spatial autoregressive model

Access to Document

10.1016/j.regsciurbeco.2012.07.005

Cite this

@article{23da78ff30ef4112866fecf98ced3e79,

title = "Model selection using J-test for the spatial autoregressive model vs. the matrix exponential spatial model",

abstract = "We consider using the J-test procedure for the non-nested model selection problem between the spatial autoregressive (SAR) model and the matrix exponential spatial specification (MESS) model. The 2SLS and GMM methods are used to implement the J-test procedure and derive several test statistics under the GMM framework. We investigate the behavior of those J-test statistics in terms of pseudo true values. We extend the J-test procedure into the setting when error terms in the model are with unknown heteroskedasticity. Monte Carlo results suggest with strong spatial dependence the J-test statistics can have good power to distinguish the SAR and MESS models.",

keywords = "GMM, J-test, Matrix exponential spatial model, Pseudo true value, Spatial autoregressive model",

author = "Xiaoyi Han and Lee, \{Lung fei\}",

note = "Publisher Copyright: {\textcopyright} 2012 Elsevier B.V.",

year = "2013",

month = mar,

day = "1",

doi = "10.1016/j.regsciurbeco.2012.07.005",

language = "English",

volume = "43",

pages = "250--271",

journal = "Regional Science and Urban Economics",

issn = "0166-0462",

publisher = "Elsevier B.V.",

number = "2",

}

TY - JOUR

T1 - Model selection using J-test for the spatial autoregressive model vs. the matrix exponential spatial model

AU - Han, Xiaoyi

AU - Lee, Lung fei

PY - 2013/3/1

Y1 - 2013/3/1

N2 - We consider using the J-test procedure for the non-nested model selection problem between the spatial autoregressive (SAR) model and the matrix exponential spatial specification (MESS) model. The 2SLS and GMM methods are used to implement the J-test procedure and derive several test statistics under the GMM framework. We investigate the behavior of those J-test statistics in terms of pseudo true values. We extend the J-test procedure into the setting when error terms in the model are with unknown heteroskedasticity. Monte Carlo results suggest with strong spatial dependence the J-test statistics can have good power to distinguish the SAR and MESS models.

AB - We consider using the J-test procedure for the non-nested model selection problem between the spatial autoregressive (SAR) model and the matrix exponential spatial specification (MESS) model. The 2SLS and GMM methods are used to implement the J-test procedure and derive several test statistics under the GMM framework. We investigate the behavior of those J-test statistics in terms of pseudo true values. We extend the J-test procedure into the setting when error terms in the model are with unknown heteroskedasticity. Monte Carlo results suggest with strong spatial dependence the J-test statistics can have good power to distinguish the SAR and MESS models.

KW - GMM

KW - J-test

KW - Matrix exponential spatial model

KW - Pseudo true value

KW - Spatial autoregressive model

UR - https://www.webofscience.com/wos/woscc/full-record/WOS:000316537100006

UR - https://www.scopus.com/pages/publications/84868236565

U2 - 10.1016/j.regsciurbeco.2012.07.005

DO - 10.1016/j.regsciurbeco.2012.07.005

M3 - Journal Article

SN - 0166-0462

VL - 43

SP - 250

EP - 271

JO - Regional Science and Urban Economics

JF - Regional Science and Urban Economics

IS - 2

ER -

Model selection using J-test for the spatial autoregressive model vs. the matrix exponential spatial model

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this