SPATIALREG Procedure

Missing values can occur in both the primary data set (specified in the DATA= option) and the secondary spatial weights data set (specified in the WMAT= option). PROC SPATIALREG does not allow missing values in the primary input data set. If any observation in the primary input data set has a missing value for one or more of the regressors or the dependent variable, PROC SPATIALREG will not fit the model. If any observation in the primary data set has a missing value for the spatial ID variable when you specify the SPATIALID statement, PROC SPATIALREG will not fit the model. In such cases, an error message is issued.

For the secondary spatial weights data set, a missing value for the spatial ID variable is not allowed when you specify the SPATIALID statement. Moreover, missing spatial weights are not allowed if you specify the NONORMALIZE option. In these cases, PROC SPATIALREG issues an error message and skips the model fitting.