is an option to SpatialEstimate and other spatial functions that gives the noise variance level in the data.
- SpatialNoiseLevel is also known as a nugget or variogram nugget.
- SpatialNoiseLevel effectively describes the data measurement error.
- SpatialNoiseLevel together with VariogramFunction is used to make local predictions of spatial values. Combining local predictions with a global trend gives the full spatial prediction function.
- The effect on spatial prediction of specifying positive SpatialNoiseLevel is that the prediction function is no longer required to interpolate through the values at each location. In such a case, the prediction surface no longer interpolates the data, but it is a smoother approximating surface.
- Possible settings for SpatialNoiseLevel:
Automatic estimate the noise level from the data θ specifiy the non-negative noise level of size
Examplesopen allclose all
Basic Examples (2)
Add noise variance to VariogramModel:
Create prediction surface with and without spatial noise variance:
Create spatial predictor function with zero noise level:
Compute the prediction function with positive noise level:
SpatialNoiseLevel can create a smoother picture of the data:
Standard ways to visualize the data:
Spatial estimate with smoothing:
Properties & Relations (1)
Only with zero SpatialNoiseLevel is SpatialEstimate an exact interpolator:
Compute spatial estimate for varying values of spatial noise level:
Wolfram Research (2021), SpatialNoiseLevel, Wolfram Language function, https://reference.wolfram.com/language/ref/SpatialNoiseLevel.html.
Wolfram Language. 2021. "SpatialNoiseLevel." Wolfram Language & System Documentation Center. Wolfram Research. https://reference.wolfram.com/language/ref/SpatialNoiseLevel.html.
Wolfram Language. (2021). SpatialNoiseLevel. Wolfram Language & System Documentation Center. Retrieved from https://reference.wolfram.com/language/ref/SpatialNoiseLevel.html