This functor iteratively solves a spatial lag regression.
| Name | Type | Description |
|---|---|---|
| P Lag | Double | pLag is the autoregressive coefficient. |
| W Neighborhoods | Neighborhood Table | The neighborhood matrix. |
| X1 | Lookup Table | x1 is the autoregressive term. When y is not known, x1 is used instead. In this case, a lookup table with same number of records and values equal to zero must be reported. |
| B Coefficients | Lookup Table | Coefficients for independent variables x2, x3, x4… xn. |
| E Error | Double | Regression random error term. |
| Name | Type | Description |
|---|---|---|
| Y Result | Lookup Table | A lookup table with Y results. |
| Y Predicted Result | Lookup Table | A lookup table with the predicted results. |
The lag spatial model is represented as follows:
where
is the autoregressive coefficient; W is the spatial weight matrix; y is the dependent variable; X is co-variables' information matrix;
is the regression coefficients and
is a random error term. W can be understood as the representation of the spatial interaction of a phenomenon. In a binary matrix, unit i is unit j’s neighbor if the spatial weight matrix cell, aij, is equal to 1.
ANSELIN, L. SpaceStat TUTORIAL. Urbana-Champaign, University of Illinois, 1992.
ANSELIN, L. Spatial Externalities, Spatial Multipliers and Spatial Econometrics. Urbana-Champaign, University of Illinois, 2002.
CalcSpatialLag