Skip to contents

Extract the Matern (theta) parameters from a PrestoGP model.

Source: R/PrestoGP_Model.R
get_theta-PrestoGPModel-method.Rd

This method extracts a list containing the Matern variance/covariance parameters from a PrestoGP model.

Usage

# S4 method for class 'PrestoGPModel'
get_theta(model)

Arguments

model

The PrestoGP model object

Value

A list containing the following elements:

sigma:

A vector containing the estimates of the variance parameter sigma for each outcome.

scale:

A vector containing the estimates of the the scale parameter(s) for each outcome.

smoothness:

A vector containing the estimates of the smoothness parameter for each outcome.

nuggets:

A vector containing the estimates of the nugget variance for each outcome.

correlation:

A matrix containing the estimates of the correlations between outcomes. Omitted for univariate models.

References

  • Messier, K.P. and Katzfuss, M. "Scalable penalized spatiotemporal land-use regression for ground-level nitrogen dioxide", The Annals of Applied Statistics (2021) 15(2):688-710.

  • Porcu, E., Bevilacqua, M., Schaback, R., and Oates RJ. "The Matérn Model: A Journey Through Statistics, Numerical Analysis and Machine Learning", Statistical Science (2024) 39(3):469-492.

See also

PrestoGPModel-class, prestogp_fit

Examples

data(soil)
soil <- soil[!is.na(soil[,5]),] # remove rows with NA's
y <- soil[,4]                   # predict moisture content
X <- as.matrix(soil[,5:9])
locs <- as.matrix(soil[,1:2])

soil.vm <- new("VecchiaModel", n_neighbors = 10)
soil.vm <- prestogp_fit(soil.vm, y, X, locs)
#> 
#> Estimating initial beta... 
#> Estimation of initial beta complete 
#> 
#> Beginning iteration 1 
#> Estimating theta... 
#> Estimation of theta complete 
#> Estimating beta... 
#> Estimation of beta complete 
#> Iteration 1 complete 
#> Current penalized negative log likelihood: 487.7036 
#> Current MSE: 9.104869 
#> Beginning iteration 2 
#> Estimating theta... 
#> Estimation of theta complete 
#> Estimating beta... 
#> Estimation of beta complete 
#> Iteration 2 complete 
#> Current penalized negative log likelihood: 482.5927 
#> Current MSE: 9.044309 
#> Beginning iteration 3 
#> Estimating theta... 
#> Estimation of theta complete 
#> Estimating beta... 
#> Estimation of beta complete 
#> Iteration 3 complete 
#> Current penalized negative log likelihood: 482.5153 
#> Current MSE: 9.04027 
#> Beginning iteration 4 
#> Estimating theta... 
#> Estimation of theta complete 
#> Estimating beta... 
#> Estimation of beta complete 
#> Iteration 4 complete 
#> Current penalized negative log likelihood: 482.5153 
#> Current MSE: 9.022014 
get_theta(soil.vm)
#> $sigma
#> [1] 10.28376
#> 
#> $scale
#> [1] 13.52572
#> 
#> $smoothness
#> [1] 0.9088149
#> 
#> $nuggets
#> [1] 0.7629514
#>