CSPATIALREG Procedure
PROC CSPATIALREG Statement
PROC CSPATIALREG options;
You can specify the following options in the PROC CSPATIALREG statement.
Data Table Options
You must specify the following option:
For all models except a purely linear model, you must also specify the following option:
- WMAT=libref.data-table
-
specifies the secondary spatial weights data table, which you can use to construct the spatial weights matrix
. libref.data-table is a two-level name, where libref refers to the library, and data-table specifies the name of the input data table. For more information about this two-level name, see the DATA= option and the section Using CAS Sessions and CAS Engine Librefs. Loosely speaking, the entries of
, 
, define the amount of influence that a unit 
has over a unit 
. The entries must be nonnegative and have zeros on the diagonal; that is, 
and 
, where 
and n is the total number of spatial units in the data. Any nonzero diagonal elements 
are replaced with 0. The spatial weights matrix can be asymmetric; that is, it is not necessary that 
. For information about missing spatial weights in , see the NONORMALIZE option. The
matrix can take two different forms: You can provide a full spatial weights matrix. In this case, the data table that you specify in the WMAT= option has n rows and
columns and must have a column for spatial ID variable. You can specify the spatial weights matrix by using a compact form when appropriate. In this form, the number of observations in the data table that you specify in the WMAT= option should match the number of nonzero elements in the spatial weights matrix. Moreover, the number of columns in this data table should be three. The first two columns contain the row and column indices for nonzero entries in the spatial weights matrix, and the third column contains the nonzero entries in the spatial weights matrix. If you use the compact form for the spatial weights matrix, you must include a SPATIALID statement to match observations in the data tables that you specify in the DATA= and WMAT= options. For more information about the SPATIALID statement, see the section SPATIALID Statement. For more information about the compact representation of the spatial weights matrix, see the section Compact Representation of the Spatial Weights Matrix.
For all models except a purely linear model, you can also specify the following option:
- NONORMALIZE
suppresses the row standardization of the spatial weights matrix that you specify in the WMAT= option. By default, the spatial weights matrix is row-standardized; that is, the spatial weights matrix has unit row sum. If you specify the NONORMALIZE option, spatial weights are used "as is" except for
, which is always treated as 0. This implies that an entry in the matrix cannot be missing for 
if you specify this option. If you do not specify this option, missing spatial weights are replaced with zeros.
Approximation Control Option
For the SAR, SDM, SEM, SDEM, and CAR models, you can specify the following options:
- APPROXIMATION=(approx-options)
-
specifies options that are related to approximating the Jacobian, as described in the section Approximations to the Jacobian. To invoke approximation, you must specify one or more of the following approx-options:
- CHEBYSHEV | TAYLOR
specifies the approximation method. By default, Chebyshev approximation is used. The Taylor approximation is used only if you specify the TAYLOR option.
- NMC=number
specifies a positive integer as the number of standard random normal draws for Monte Carlo simulation. By default, NMC=100.
- ORDER=number
specifies a positive integer as the order of series in Taylor or Chebyshev approximation. If you specify Taylor approximation, ORDER=50 by default. If you specify Chebyshev approximation, ORDER=5 by default.
- SEED=number
specifies an integer seed in the range 1 to
for the random number generator that is used for Monte Carlo simulation. Specifying a seed enables you to reproduce your analysis. By default, SEED=1.
Impact Estimation Control Option
For impact estimation, you can specify the following options:
- IMPACT(impactest-option)
-
specifies options that are related to impact estimation, as described in the section Impact Estimation. To invoke impact estimation, you must specify one or more of the following impactest-options:
- NMC=number
specifies a positive integer as the number of iterations for Monte Carlo simulation. By default, NMC=1000.
- ORDER=number
specifies a positive integer as the highest order of the Neumann series for approximation. By default, ORDER=50.
- SEED=number
specifies an integer seed in the range 1 to
for the random number generator that is used for Monte Carlo simulation. Specifying a seed enables you to reproduce your analysis. By default, SEED=1.
You can use the IMPACT option to summarize the average direct impacts, the average indirect impacts, and the average total impacts of explanatory variables in the model.
Printing Options
You can specify the following options in either the PROC CSPATIALREG statement or the MODEL statement:
Estimation Control Options
You can specify the following options in either the PROC CSPATIALREG statement or the MODEL statement:
Optimization Control Options
PROC CSPATIALREG uses the nonlinear optimization (NLO) subsystem to perform nonlinear optimization tasks. You can specify the following options in either the PROC CSPATIALREG statement or the MODEL statement: