alik_cutoff: Approximate log-likelihood calculation

In geoBayes: Analysis of Geostatistical Data using Bayes and Empirical Bayes Methods

Approximate log-likelihood calculation

Description

Calculate the likelihood approximation at different parameter values. This function is useful for choosing the skeleton set.

Plot likelihood approximation.

Usage

alik_cutoff(likopt, par_vals, likthreshold)

alik_plot(alikobj)

Arguments

likopt	Output from the function alik_optim.
par_vals	A named list with some of the components "linkp", "phi", "omg", "kappa".
likthreshold	A threshold value proportion to calculate the cutoff. The cutoff will be calculated as that proportion relative to the maximum value of the log-likelihood.
alikobj	Output from alik_cutoff.

Details

The input par_vals is meant to contain vector of parameter values for each parameter. For each element in par_vals, the other parameters are set equal to the maximisers given in likopt and the approximate likelihood is computed. The cuttoff is calculated using linear interpolation provided by approx.

The plot can be used to visualise the Laplace approximation to the likelihood provided by the function alik_cutoff.

Value

A list with the log-likelihood approximation and cutoff values.

Draws a plot.

References

Evangelou, E., & Roy, V. (2019). Estimation and prediction for spatial generalized linear mixed models with parametric links via reparameterized importance sampling. Spatial Statistics, 29, 289-315.

geoBayes documentation built on Aug. 21, 2023, 9:08 a.m.