Tool GWR for Multiple Predictor Grids

Geographically Weighted Regression for a multiple predictors supplied as grids, to which the regression model is applied. Further details can be stored optionally.

References

Fotheringham, S.A., Brunsdon, C., Charlton, M. (2002): Geographically Weighted Regression: the analysis of spatially varying relationships. John Wiley & Sons. online.
Fotheringham, S.A., Charlton, M., Brunsdon, C. (1998): Geographically weighted regression: a natural evolution of the expansion method for spatial data analysis. online.
Lloyd, C. (2010): Spatial Data Analysis - An Introduction for GIS Users. Oxford, 206p.
Zhang, D., Ren, N., and Hou, X. (2018): An improved logistic regression model based on a spatially weighted technique (ILRBSWT v1.0) and its application to mineral prospectivity mapping. Geosci. Model Dev., 11, 2525-2539. doi:10.5194/gmd-11-2525-2018.

Author: O.Conrad (c) 2010
Menu: Spatial and Geostatistics|Geographically Weighted Regression

Parameters

	Name	Type	Identifier	Description	Constraints
Input	Points	Shapes (input)	`POINTS`	-	-
Input	Predictors	Grid list (input)	`PREDICTORS`	-	-
Output	Residuals (*)	Shapes (optional output)	`RESIDUALS`	-	-
	Regression	Grid (output)	`REGRESSION`	-	-
	Coefficient of Determination	Grid (output)	`QUALITY`	-	-
	Regression Parameters (*)	Grid list (optional output)	`MODEL`	-	-
Options	Grid system	Grid system	`PARAMETERS_GRID_SYSTEM`	-	-
	Dependent Variable	Table field	`DEPENDENT`	-	-
	Logistic Regression	Boolean	`LOGISTIC`	-	Default: 0
	Output of Regression Parameters	Boolean	`MODEL_OUT`	-	Default: 0
	Model Resolution	Choice	`RESOLUTION`	-	Available Choices: [0] same as predictors [1] user defined Default: 1
	Resolution	Floating point	`RESOLUTION_VAL`	map units	Minimum: 0.000000 Default: 1.000000
	Weighting Function	Choice	`DW_WEIGHTING`	-	Available Choices: [0] no distance weighting [1] inverse distance to a power [2] exponential [3] gaussian weighting Default: 3
	Inverse Distance Weighting Power	Floating point	`DW_IDW_POWER`	-	Minimum: 0.000000 Default: 1.000000
	Inverse Distance Offset	Boolean	`DW_IDW_OFFSET`	Calculates weights for distance plus one, avoiding division by zero for zero distances	Default: 1
	Gaussian and Exponential Weighting Bandwidth	Floating point	`DW_BANDWIDTH`	-	Minimum: 0.000000 Default: 1.000000
	Search Range	Choice	`SEARCH_RANGE`	-	Available Choices: [0] local [1] global Default: 1
	Maximum Search Distance	Floating point	`SEARCH_RADIUS`	local maximum search distance given in map units	Minimum: 0.000000 Default: 1000.000000
	Number of Points	Choice	`SEARCH_POINTS_ALL`	-	Available Choices: [0] maximum number of nearest points [1] all points within search distance Default: 1
	Minimum	Integer	`SEARCH_POINTS_MIN`	minimum number of points to use	Minimum: 1 Default: 16
	Maximum	Integer	`SEARCH_POINTS_MAX`	maximum number of nearest points	Minimum: 1 Default: 20
	Direction	Choice	`SEARCH_DIRECTION`	-	Available Choices: [0] all directions [1] quadrants Default: 0
(*) optional

Command-line

Usage: saga_cmd statistics_regression 6 [-POINTS <str>] [-DEPENDENT <str>] [-RESIDUALS <str>] [-PREDICTORS <str>] [-REGRESSION <str>] [-QUALITY <str>] [-MODEL <str>] [-LOGISTIC <str>] [-MODEL_OUT <str>] [-RESOLUTION <str>] [-RESOLUTION_VAL <double>] [-DW_WEIGHTING <str>] [-DW_IDW_POWER <double>] [-DW_IDW_OFFSET <str>] [-DW_BANDWIDTH <double>] [-SEARCH_RANGE <str>] [-SEARCH_RADIUS <double>] [-SEARCH_POINTS_ALL <str>] [-SEARCH_POINTS_MIN <num>] [-SEARCH_POINTS_MAX <num>] [-SEARCH_DIRECTION <str>]
  -POINTS:<str>           	Points
	Shapes (input)
  -DEPENDENT:<str>        	Dependent Variable
	Table field
  -RESIDUALS:<str>        	Residuals
	Shapes (optional output)
  -PREDICTORS:<str>       	Predictors
	Grid list (input)
  -REGRESSION:<str>       	Regression
	Grid (output)
  -QUALITY:<str>          	Coefficient of Determination
	Grid (output)
  -MODEL:<str>            	Regression Parameters
	Grid list (optional output)
  -LOGISTIC:<str>         	Logistic Regression
	Boolean
	Default: 0
  -MODEL_OUT:<str>        	Output of Regression Parameters
	Boolean
	Default: 0
  -RESOLUTION:<str>       	Model Resolution
	Choice
	Available Choices:
	[0] same as predictors
	[1] user defined
	Default: 1
  -RESOLUTION_VAL:<double>	Resolution
	Floating point
	Minimum: 0.000000
	Default: 1.000000
  -DW_WEIGHTING:<str>     	Weighting Function
	Choice
	Available Choices:
	[0] no distance weighting
	[1] inverse distance to a power
	[2] exponential
	[3] gaussian weighting
	Default: 3
  -DW_IDW_POWER:<double>  	Inverse Distance Weighting Power
	Floating point
	Minimum: 0.000000
	Default: 1.000000
  -DW_IDW_OFFSET:<str>    	Inverse Distance Offset
	Boolean
	Default: 1
  -DW_BANDWIDTH:<double>  	Gaussian and Exponential Weighting Bandwidth
	Floating point
	Minimum: 0.000000
	Default: 1.000000
  -SEARCH_RANGE:<str>     	Search Range
	Choice
	Available Choices:
	[0] local
	[1] global
	Default: 1
  -SEARCH_RADIUS:<double> 	Maximum Search Distance
	Floating point
	Minimum: 0.000000
	Default: 1000.000000
  -SEARCH_POINTS_ALL:<str>	Number of Points
	Choice
	Available Choices:
	[0] maximum number of nearest points
	[1] all points within search distance
	Default: 1
  -SEARCH_POINTS_MIN:<num>	Minimum
	Integer
	Minimum: 1
	Default: 16
  -SEARCH_POINTS_MAX:<num>	Maximum
	Integer
	Minimum: 1
	Default: 20
  -SEARCH_DIRECTION:<str> 	Direction
	Choice
	Available Choices:
	[0] all directions
	[1] quadrants
	Default: 0