Overview: VARIOGRAM Procedure

The VARIOGRAM procedure computes empirical measures of spatial continuity for two-dimensional spatial data. These measures are a function of the distances between the sample data pairs. When the data are free of nonrandom (or systematic) surface trends, the estimated continuity measures are the empirical semivariance and covariance. The procedure also fits permissible theoretical models to the empirical semivariograms, so that you can use them in subsequent analysis to perform spatial prediction. You can produce plots of the empirical semivariograms in addition to plots of the fitted models. Both isotropic and anisotropic continuity measures are available.

PROC VARIOGRAM also provides the Moran’s and Geary’s spatial autocorrelation statistics, in addition to the Moran scatter plot to visualize spatial associations within a specified neighborhood around observations. The procedure produces the OUTVAR=, OUTPAIR=, and OUTDISTANCE= data sets that contain information about the semivariogram analysis. Also, the OUTACWEIGHTS= and the OUTMORAN= output data sets contain information about the autocorrelation analysis.

The VARIOGRAM procedure now uses ODS Graphics to create graphs as part of its output. For general information about ODS Graphics, see Chapter 21, Statistical Graphics Using ODS. For more information about the graphics available in PROC VARIOGRAM, see the section ODS Graphics.