Measurement errors in a spatial context

J. Le Gallo, B. Fingleton

Research output: Contribution to journalConference Contribution

5 Citations (Scopus)

Abstract

Measurement error in an independent variable is one reason why OLS estimates may not be consistent. However, as shown by Dagenais (1994), in some circumstances the OLS bias may be ameliorated somewhat given the presence of serially correlated disturbances, and OLS may prove superior to standard techniques used to correct for serial correlation. This paper considers the case of cross-sectional regression models with measurement errors in the explanatory variables and with spatial dependence. The study focuses on the evidence provided by an empirical illustration and Monte Carlo experiments examining measurement error impact in the presence of autoregressive error processes and autoregressive spatial lags.
LanguageEnglish
Pages114-125
Number of pages12
JournalRegional Science and Urban Economics
Volume42
Issue number1-2
DOIs
Publication statusPublished - 2012

Fingerprint

disturbance
experiment
regression
Measurement error
trend
evidence
Monte Carlo experiment
Regression model
Spatial dependence
Serial correlation
Spatial lag
Cross-sectional regression

Keywords

  • autoregressive model
  • Monte-Carlo simulations
  • measurement error
  • spatial autocorrelation
  • matrices
  • instrumental variables

Cite this

Le Gallo, J. ; Fingleton, B. / Measurement errors in a spatial context. In: Regional Science and Urban Economics. 2012 ; Vol. 42, No. 1-2. pp. 114-125.
@article{05ef7de049254434b2e00baa675b702b,
title = "Measurement errors in a spatial context",
abstract = "Measurement error in an independent variable is one reason why OLS estimates may not be consistent. However, as shown by Dagenais (1994), in some circumstances the OLS bias may be ameliorated somewhat given the presence of serially correlated disturbances, and OLS may prove superior to standard techniques used to correct for serial correlation. This paper considers the case of cross-sectional regression models with measurement errors in the explanatory variables and with spatial dependence. The study focuses on the evidence provided by an empirical illustration and Monte Carlo experiments examining measurement error impact in the presence of autoregressive error processes and autoregressive spatial lags.",
keywords = "autoregressive model, Monte-Carlo simulations, measurement error, spatial autocorrelation, matrices, instrumental variables",
author = "{Le Gallo}, J. and B. Fingleton",
year = "2012",
doi = "10.1016/j.regsciurbeco.2011.08.004",
language = "English",
volume = "42",
pages = "114--125",
journal = "Regional Science and Urban Economics",
issn = "0166-0462",
number = "1-2",

}

Measurement errors in a spatial context. / Le Gallo, J.; Fingleton, B.

In: Regional Science and Urban Economics, Vol. 42, No. 1-2, 2012, p. 114-125.

Research output: Contribution to journalConference Contribution

TY - JOUR

T1 - Measurement errors in a spatial context

AU - Le Gallo, J.

AU - Fingleton, B.

PY - 2012

Y1 - 2012

N2 - Measurement error in an independent variable is one reason why OLS estimates may not be consistent. However, as shown by Dagenais (1994), in some circumstances the OLS bias may be ameliorated somewhat given the presence of serially correlated disturbances, and OLS may prove superior to standard techniques used to correct for serial correlation. This paper considers the case of cross-sectional regression models with measurement errors in the explanatory variables and with spatial dependence. The study focuses on the evidence provided by an empirical illustration and Monte Carlo experiments examining measurement error impact in the presence of autoregressive error processes and autoregressive spatial lags.

AB - Measurement error in an independent variable is one reason why OLS estimates may not be consistent. However, as shown by Dagenais (1994), in some circumstances the OLS bias may be ameliorated somewhat given the presence of serially correlated disturbances, and OLS may prove superior to standard techniques used to correct for serial correlation. This paper considers the case of cross-sectional regression models with measurement errors in the explanatory variables and with spatial dependence. The study focuses on the evidence provided by an empirical illustration and Monte Carlo experiments examining measurement error impact in the presence of autoregressive error processes and autoregressive spatial lags.

KW - autoregressive model

KW - Monte-Carlo simulations

KW - measurement error

KW - spatial autocorrelation

KW - matrices

KW - instrumental variables

UR - http://www.scopus.com/inward/record.url?scp=80052529448&partnerID=8YFLogxK

U2 - 10.1016/j.regsciurbeco.2011.08.004

DO - 10.1016/j.regsciurbeco.2011.08.004

M3 - Conference Contribution

VL - 42

SP - 114

EP - 125

JO - Regional Science and Urban Economics

T2 - Regional Science and Urban Economics

JF - Regional Science and Urban Economics

SN - 0166-0462

IS - 1-2

ER -