This documentation is for a version of the software that is not covered by Standard Support. Select a different version from the version selector in the banner, or access the latest documentation.
The MODEL statement specifies the dependent-variable and independent covariates (regressors) for the regression model. If you specify no regressors, PROC CSPATIALREG fits a model that contains only an intercept. The dependent-variable is treated as a continuous variable in the primary input data table (specified in the DATA= option). Models in PROC CSPATIALREG do not allow missing values. If there are missing values, an error message is returned.
You can specify only one MODEL statement. You can specify the following options after a slash (/):
NOINT
suppresses the intercept parameter.
TYPE=ALL | AUTO | CAR | LINEAR | SAC | SAR | SARMA | SEM | SMA
specifies the type of model to be fit. If you specify this option in both the MODEL statement and the PROC CSPATIALREG statement, the MODEL statement overrides the PROC CSPATIALREG statement. You can specify the following values:
ALL
fits all models.
AUTO
fits a subset of selected models.
CAR
fits a conditional autoregressive model.
LINEAR
fits a linear model.
SAC
fits a spatial autoregressive confused model.
SAR
fits a spatial autoregressive model.
SARMA
fits a spatial autoregressive moving average model.
SEM
fits a spatial error model.
SMA
fits a spatial moving average model.
By default, TYPE=SAR. For more information about the TYPE=ALL or TYPE=AUTO option, see the section Multiple Model Comparison.
Printing Options
You can specify the following options in either the PROC CSPATIALREG statement or the MODEL statement. If you specify these options in both the PROC CSPATIALREG and MODEL statements, the MODEL statement overrides the PROC CSPATIALREG statement.
CORRB
prints the correlation matrix of the parameter estimates.
COVB
prints the covariance matrix of the parameter estimates.