GWR for Grid Downscaling

Author: O.Conrad (c) 2013
Menu: Spatial and Geostatistics | Geographically Weighted Regression

Description

Geographically Weighted Regression for grid downscaling.

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. Environment and Planning A 30(11), 1905–1927. 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.

Parameters

	Name	Type	Identifier	Description	Constraints
Input	Predictors	grid list, input	`PREDICTORS`	-	-
Input	Dependent Variable	grid, input	`DEPENDENT`	-	-
Output	Regression	grid, output	`REGRESSION`	-	-
	Regression with Residual Correction	grid, output, optional	`REG_RESCORR`	-	-
	Coefficient of Determination	grid, output	`QUALITY`	-	-
	Residuals	grid, output	`RESIDUALS`	-	-
	Regression Parameters	grid list, output, optional	`MODEL`	-	-
Options	Grid System	grid system	`PARAMETERS_GRID_SYSTEM`	-	-
	Grid System	grid system	`GRID_SYSTEM`	-	-
	Logistic Regression	boolean	`LOGISTIC`	-	Default: 0
	Output of Model Parameters	boolean	`MODEL_OUT`	-	Default: 0
	Search Range	choice	`SEARCH_RANGE`	-	Available Choices: [0] local [1] global Default: 0
	Search Distance [Cells]	integer number	`SEARCH_RADIUS`	-	Minimum: 1 Default: 10
	Weighting Function	choice	`DW_WEIGHTING`	-	Available Choices: [0] no distance weighting [1] inverse distance to a power [2] exponential [3] gaussian Default: 3
	Power	floating point number	`DW_IDW_POWER`	-	Minimum: 0.000000 Default: 2.000000
	Bandwidth	floating point number	`DW_BANDWIDTH`	Bandwidth for exponential and Gaussian weighting	Minimum: 0.000000 Default: 7.000000

Command Line


Usage: saga_cmd statistics_regression 14 [-PREDICTORS ] [-REGRESSION ] [-REG_RESCORR ] [-DEPENDENT ] [-QUALITY ] [-RESIDUALS ] [-MODEL ] [-LOGISTIC ] [-MODEL_OUT ] [-SEARCH_RANGE ] [-SEARCH_RADIUS ] [-DW_WEIGHTING ] [-DW_IDW_POWER ] [-DW_BANDWIDTH ]
  -PREDICTORS:        Predictors
	grid list, input
  -REGRESSION:        Regression
	grid, output
  -REG_RESCORR:       Regression with Residual Correction
	grid, output, optional
  -DEPENDENT:         Dependent Variable
	grid, input
  -QUALITY:           Coefficient of Determination
	grid, output
  -RESIDUALS:         Residuals
	grid, output
  -MODEL:             Regression Parameters
	grid list, output, optional
  -LOGISTIC:          Logistic Regression
	boolean
	Default: 0
  -MODEL_OUT:         Output of Model Parameters
	boolean
	Default: 0
  -SEARCH_RANGE:      Search Range
	choice
	Available Choices:
	[0] local
	[1] global
	Default: 0
  -SEARCH_RADIUS:     Search Distance [Cells]
	integer number
	Minimum: 1
	Default: 10
  -DW_WEIGHTING:
                           Weighting Function
                           	choice
                           	Available Choices:
                           	[0] no distance weighting
                           	[1] inverse distance to a power
                           	[2] exponential
                           	[3] gaussian
                           	Default: 3
  -DW_IDW_POWER:   Power
	floating point number
	Minimum: 0.000000
	Default: 2.000000
  -DW_BANDWIDTH:   Bandwidth
	floating point number
	Minimum: 0.000000
	Default: 7.000000