Geographically weighted regression (GWR) is a useful technique for exploring spatial nonstationarity by calibrating, for example, a regression model which allows different relationships to exist at different points in space. In this line of research, many spatial data sets have been successfully analyzed and some statistical tests for spatial variation have been developed. However, an important assumption in these studies is that the disturbance terms of the GWR model are uncorrelated and of common variance. Similar to the case in the ordinary linear regression, spatial autocorrelation can invalidate the standard assumption of homoscedasticity of the disturbances and mislead the results of statistical inference. Therefore, developing some statistical methods to test for spatial autocorrelation is a very important issue. In this paper, two kinds of the statistical tests for spatial autocorrelation among the residuals of the GWR model are suggested. Also, an efficient approximation method for calculating the p-values of the test statistics is proposed. Some simulations are run to examine the performances of the proposed methods and the results are encouraging. The study not only makes it possible to test for spatial autocorrelation among the GWR residuals in a conventional statistical manner, but also provides a useful means for model validation.
Get full access to this article
View all access options for this article.
References
1.
AnselinL, 1988aSpatial Econometrics: Methods and Models (Kluwer Academic, Dordrecht)
2.
AnselinL, 1988b, “Lagrange multiplier test diagnostics for spatial dependence and spatial heterogeneity”Geographical Analysis201–17
3.
AnselinL, 1990, “Spatial dependence and spatial structural instability in applied regression analysis”Journal of Regional Science30185–207
4.
AnselinL, 1995, “Local indicators of spatial association—LISA”Geographical Analysis2791–115
5.
AnselinLGriffithD A, 1988, “Do spatial effects really matter in regression analysis?”Papers of the Regional Science Association6511–34
6.
AnselinLReyS, 1991, “Properties of tests for spatial dependence in linear regression models”Geographical Analysis23112–131
7.
AzzaliniABowmanA, 1993, “On the use of nonparametric regression for checking linear relationships”Journal of the Royal Statistical Society B55549–557
8.
BowmanA W, 1984, “An alternative method of cross-validation for the smoothing of density estimate”Biometrika71353–360
9.
BrunsdonCFotheringhamA SCharltonM, 1996, “Geographically weighted regression: a method for exploring spatial nonstationarity”Geographical Analysis28281–298
10.
BrunsdonCFotheringhamA SCharltonM, 1997, “Geographical instability in linear regression modelling—a preliminary investigation”, in New Techniques and Technologies for Statistics II (IOS Press, Amsterdam) pp 149–158
11.
BrunsdonCFotheringhamA SCharltonM, 1999, “Some notes on parametric significance test for geographically weighted regression”Journal of Regional Science39497–524
12.
BurridgeP, 1980, “On the Cliff–Ord test for spatial correlation”Journal of the Royal Statistical Society B42107–108
13.
CasettiE, 1972, “Generating models by the expansion method: applications to geographical research”Geographical Analysis481–91
14.
CasettiE, 1986, “The dual expansion method: an application for evaluating the effects of population growth on development”IEEE Transactions on Systems, Man and Cybernetics1629–39
15.
ClevelandW S, 1979, “Robust locally weighted regression and smoothing scatterplots”Journal of the American Statistical Association74829–836
16.
ClevelandW SDevlinS J, 1988, “Locally weighted regression: an approach to regression analysis by local fitting”Journal of the American Statistical Association83596–610
17.
ClevelandW SDevlinS JGrosseE, 1988, “Regression by local fitting: methods, properties and computational algorithms”Journal of Econometrics3787–114
18.
CliffA DOrdJ K, 1972, “Testing for spatial autocorrelation among regression residuals”Geographical Analysis4267–284
CliffA DOrdJ K, 1981Spatial Processes, Models and Applications (Pion, London)
21.
CordyC BGriffithD A, 1993, “Efficiency of least squares estimators in the presence of spatial autocorrelation”Communications in Statistics: Simulation and Computation221161–1179
22.
DurbinJWatsonG S, 1950, “Testing for serial correlation in least squares regression I”Biometrika37409–428
23.
EubankR LSpiegelmanC H, 1990, “Testing the goodness of fit of a linear model via nonparametric regression techniques”Journal of the American Statistical Association85387–392
24.
FosterS AGorrW L, 1986, “An adaptive filter for estimating spatially-varying parameters: application to modelling police hours spent in response to calls for service”Management Science32878–889
25.
FotheringhamA SPittsT C, 1995, “Directional variation in distance-decay”Environment and Planning A27715–729
26.
FotheringhamA SCharltonMBrunsdonC, 1997a, “Measuring spatial variations in relationships with geographically weighted regression”, in Recent Developments in Spatial Analysis Eds FischerM MGetisA (Springer, London) pp 60–82
27.
FotheringhamA SCharltonMBrunsdonC, 1997b, “Two techniques for exploring non-stationarity in geographical data”Geographical Systems459–82
28.
GearyR C, 1954, “The contiguity ratio and statistical mapping”The Incorporated Statistician5115–145
29.
GorrW LOlligschlaegerA M, 1994, “Weighted spatial adaptive filtering: Monte Carlo studies and application to illicit drug market modelling”Geographical Analysis2667–87
30.
GriffithD A, 1988Advanced Spatial Statistics (Kluwer Academic, Dordrecht)
31.
HärdleWMammenE, 1993, “Comparing nonparametric versus parametric regression fits”The Annals of Statistics211926–1947
32.
HeppleL W, 1998, “Exact testing for spatial autocorrelation among regression residuals”Environment and Planning A3085–109
33.
HockingR R, 1996Methods and Applications of Linear Models (John Wiley, New York)
34.
ImhofJ P, 1961, “Computing the distribution of quadratic forms in normal variables”Biometrika48419–426
35.
JayasuriyaR, 1996, “Testing for polynomial regression using non-parametric regression techniques”Journal of the American Statistical Association911626–1631
36.
JohnsonN L, 1959, “On an extension of the connection between Poisson and χ2 distributions”Biometrika46352–363
37.
JohnsonSKauJ, 1980, “Urban spatial structure: an analysis with a varying coefficient model”Journal of Urban Economics7141–154
38.
JonesJ PCasettiE, 1992Applications of the Expansion Method (Routledge, London)
39.
KauJLeeC, 1977, “A random coefficient model to estimate a stochastic density gradient”Regional Science and Urban Economics7169–177
40.
KoertsJAbrahamseA P I, 1968, “On the power of the BLUS procedure”Journal of the American Statistical Association631227–1236
41.
KrämerWDonningerC, 1987, “Spatial autocorrelation among errors and the relative efficiency of OLS in the linear regression model”Journal of the American Statistical Association82577–579
42.
LeungYMeiC LZhangW X, 2000, “Statistical tests for spatial nonstationarity based on the geographically weighted regression model”Environment and Planning A329–32
43.
MoranP A P, 1950, “Notes on continuous stochastic phenomena”Biometrika3717–23
44.
NeterJKutnerM HNachtsheimC JWassermanW, 1996Applied Linear Regression Models4th edition (Irwin, Homewood, IL)
45.
PearsonE S, 1959, “Note on an approximation to the distribution of non-central χ2”Biometrika46364
46.
SolomonHStephensM A, 1978, “Approximation to density functions using Pearson curves”Journal of the American Statistical Association73153–160
47.
TiefelsdorfMBootsB, 1995, “The exact distribution of Moran's I”Environment and Planning A27985–999
48.
TiefelsdorfMBootsB, 1996, “Letter to the editor: the exact distribution of Moran's I”Environment and Planning A281900
49.
WandM PJonesM C, 1995Kernel Smoothing (Chapman and Hall, London)
