Testing Regression Residuals for Spatial Autocorrelation Using SAS: a Technical Note

  • John Paul Jones III University of Kentucky
  • Stuart A. Foster Western Kentucky University
Keywords: Linear Model, Spatial Autocorrelation, Regression Residuals, Moran's I statistic, SAS package, PROC MATRIX

Abstract

Spatial autocorrelation of regression residuals is a violation of an assumption of the general linear model. Yet despite the widespread use of spatial data in applied work, many researchers fail to examine residuals for spatial autocorrelation, or they rely on subjective interpretation of residual maps. In part, this problem stems from a lack of easy-to-use programs that can be incorporated into existing research procedures. In this note we present a short program for testing spatial association among regression residuals. Based on Moran's I statistic, the program employs SAS's PROC MATRIX language, and can easily augment regression analyses run using the SAS package.

References

Cliff, A.D. and Ord, J.K. (1973). Spatial Autocorrelation. London: Pion.

Cliff, A.D. and Ord, J.K. (1980). Spatial Processes: Models and Applications. London: Pion.

Draper, N.R. and Smith, H. (1966). Applied Regression Analysis. New York: Wiley.

Gatrell, A.C. (1979). Autocorrelation in space. Environment and Planning, 11:507-516.

Poole, M.A. and O'Farrell, P.N. (1971). The assumptions of the linear regression model. Transactions of the Institute of British Geographers, 52:145-156.

SAS User's Guide,: Basics (1982a). Cary, NC: SAS Institute.

SAS User's Guide: 'statistics (1982b). Cary, NC: SAS Institute.

Published
2016-02-13
Issue
Vol 11 (1991): Geography Research Forum
Section
Articles
Authors who publish with this journal agree to the following terms:

    1. The contributor(s) (authors) warrant that the entire work is original and unpublished; it is submitted only to this Journal and all text, data, figures/tables or other illustrations included in this work are completely original and unpublished, and these have not been previously published or submitted elsewhere in any form or media whatsoever.
    2. The contributor(s) warrant that the work contains no unlawful or libelous statements and opinions and liable materials of any kind whatsoever, does not infringe on any copyrights, intellectual property rights, personal rights or rights of any kind of others, nor contains any plagiarized, fraudulent, improperly attributed materials, instructions, procedures, information or ideas that might cause any harm, damage, injury, losses or costs of any kind to person or property.
    3. The contributor(s) retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
    4. The contributor(s) are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
    5. The contributor(s) are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
    6. Geography Research Forum may disseminate the content of the publications and publications’ Meta data in text, image, or other print and electronic formats to providers of research databases (e.g. EBSCO, GeoBase, JSTOR) to facilitate publications' exposure.