SpatialTrendFunction
is an option to SpatialEstimate that specifies what global trend model to use for data.
Details
- SpatialTrendFunction specifies the family of global trends to fit to data. The global trend is a part of the spatial prediction model, where the remaining parts come from local variation and noise.
- The following settings can be used:
-
Automatic automatically determine trend model n a polynomial with total degree n (universal kriging) {{f1,f2,…},{x,y,…}} estimate the model β0+β1f1+β2f2+… with the basis functions fi in the variables x,y,… as in LinearModelFit (universal kriging)
Examples
open allclose allBasic Examples (1)
Scope (3)
Properties & Relations (2)
For large spatial noise levels, the spatial estimator converges to the trend function:
Compute SpatialEstimate for increasing values of noise level for specified polynomial trend order:
The method for estimating the trend will change depending on whether we supply the variogram:
Compute spatial estimate with a given variogram model:
Compute spatial estimate with the variogram being estimated automatically:
Text
Wolfram Research (2021), SpatialTrendFunction, Wolfram Language function, https://reference.wolfram.com/language/ref/SpatialTrendFunction.html.
CMS
Wolfram Language. 2021. "SpatialTrendFunction." Wolfram Language & System Documentation Center. Wolfram Research. https://reference.wolfram.com/language/ref/SpatialTrendFunction.html.
APA
Wolfram Language. (2021). SpatialTrendFunction. Wolfram Language & System Documentation Center. Retrieved from https://reference.wolfram.com/language/ref/SpatialTrendFunction.html