sumsmoothlone

smoothed-L1 penalization over the covariate-driven parameters

Estimation, prediction, and simulation of nonstationary Gaussian process with modular covariate-based covariance functions.
Sources of nonstationarity, such as spatial mean, variance, geometric anisotropy, smoothness, and nugget, can be considered based on spatial characteristics.
An induced compact-supported nonstationary covariance function is provided, enabling fast and memory-efficient computations when handling densely sampled domains.

Federico Blasi

cocons

Covariate-Based Covariance Functions for Nonstationary Spatial
Modeling

Reinhard Furrer

sumsmoothlone function

<dl><dt>x</dt>
<dd>a vector of parameters to penalize</dd>
<dt>lambda</dt>
<dd>penalization hyperparameter</dd>
<dt>alpha</dt>
<dd>hyperparameter to approximate L1 penalization. By default 1e6 provides a decent approximation to L1, while also being a differentiable function</dd></dl>

Arguments

smoothed-L1 penalization over the covariate-driven parameters — sumsmoothlone

<dl>

<dt>x</dt>
<dd>a vector of parameters to penalize</dd>


<dt>lambda</dt>
<dd>penalization hyperparameter</dd>


<dt>alpha</dt>
<dd>hyperparameter to approximate L1 penalization. By default 1e6 provides a decent approximation to L1, while also being a differentiable function</dd>

</dl>

sumsmoothlone: smoothed-L1 penalization over the covariate-driven parameters

Description

Usage

Value

Arguments