Testing for Spatial Autocorrelation among the Residuals of the Geographically Weighted Regression

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.

[1]  Daniel A. Griffith,et al.  Efficiency of least squares estimators in the presence of spatial autocorrelation , 1993 .

[2]  W. Cleveland Robust Locally Weighted Regression and Smoothing Scatterplots , 1979 .

[3]  E. Casetti,et al.  Applications of the Expansion Method , 1991 .

[4]  J. Durbin,et al.  Testing for serial correlation in least squares regression. I. , 1950, Biometrika.

[5]  Martin Charlton,et al.  Two techniques for exploring non-stationarity in geographical data , 1997 .

[6]  J. Kau,et al.  A random coefficient model to estimate a stochastic density gradient , 1977 .

[7]  Daniel A. Griffith,et al.  Advanced Spatial Statistics , 1988 .

[8]  Luc Anselin,et al.  Do spatial effects really matter in regression analysis , 2005 .

[9]  Clifford H. Spiegelman,et al.  Testing the Goodness of Fit of a Linear Model via Nonparametric Regression Techniques , 1990 .

[10]  Matthew P. Wand,et al.  Kernel Smoothing , 1995 .

[11]  L. Anselin SPATIAL DEPENDENCE AND SPATIAL STRUCTURAL INSTABILITY IN APPLIED REGRESSION ANALYSIS , 1990 .

[12]  Bodhini R. Jayasuriya,et al.  Testing for Polynomial Regression Using Nonparametric Regression Techniques , 1996 .

[13]  R. Geary,et al.  The Contiguity Ratio and Statistical Mapping , 1954 .

[14]  Michael Tiefelsdorf,et al.  The Exact Distribution of Moran's I , 1995 .

[15]  Yee Leung,et al.  Statistical Tests for Spatial Nonstationarity Based on the Geographically Weighted Regression Model , 2000 .

[16]  Stanley R. Johnson,et al.  Urban spatial structure : an analysis with a varying coefficient model / BEBR No. 339 , 1976 .

[17]  W. Cleveland,et al.  Regression by local fitting: Methods, properties, and computational algorithms , 1988 .

[18]  W. Cleveland,et al.  Locally Weighted Regression: An Approach to Regression Analysis by Local Fitting , 1988 .

[19]  E. Mammen,et al.  Comparing Nonparametric Versus Parametric Regression Fits , 1993 .

[20]  P. Burridge,et al.  On the Cliff‐Ord Test for Spatial Correlation , 1980 .

[21]  A. Fotheringham,et al.  Directional Variation in Distance Decay , 1995 .

[22]  L. Anselin Spatial Econometrics: Methods and Models , 1988 .

[23]  A. Bowman An alternative method of cross-validation for the smoothing of density estimates , 1984 .

[24]  J. Durbin,et al.  Testing for serial correlation in least squares regression. II. , 1950, Biometrika.

[25]  J. Imhof Computing the distribution of quadratic forms in normal variables , 1961 .

[26]  Wilpen L. Gorr,et al.  An adaptive filter for estimating spatially-varying parameters: application to modeling police hours spent in response to calls for service , 1986 .

[27]  Walter Krämer,et al.  Spatial Autocorrelation among Errors and the Relative Efficiency of OLS in the Linear Regression Model , 1987 .

[28]  L. Amelin,et al.  Local Indicators of Spatial Association-LISA , 1995 .

[29]  A. P. J. Abrahamse,et al.  On the Power of the Blus Procedure , 1968 .

[30]  Adrian Bowman,et al.  On the Use of Nonparametric Regression for Checking Linear Relationships , 1993 .

[31]  Herbert Solomon,et al.  Approximations to Density Functions Using Pearson Curves , 1978 .

[32]  S. Fotheringham,et al.  Some Notes on Parametric Significance Tests for Geographically Weighted Regression , 1999 .

[33]  E. S. Pearson Note on an Approximation to the Distribution of Non-Central χ 2 , 1959 .

[34]  Angela M. Dean Methods and Applications of Linear Models , 1996 .

[35]  Lw Hepple,et al.  Exact Testing for Spatial Autocorrelation among Regression Residuals , 1998 .

[36]  N. L. Johnson ON AN EXTENSION OF THE CONNEXION BETWEEN POISSON AND X2 DISTRIBUTIONS , 1959 .

[37]  P. Moran Notes on continuous stochastic phenomena. , 1950, Biometrika.

[38]  Martin Charlton,et al.  Measuring Spatial Variations in Relationships with Geographically Weighted Regression , 1997 .

[39]  Emilio Casetti,et al.  The Dual Expansion Method: An Application for Evaluating the Effects of Population Growth on Development , 1986, IEEE Transactions on Systems, Man, and Cybernetics.