refit_inla: Fitting a modified model in INLA

In hunansona/ed4bhm: Empirical Determinacy for Bayesian Hierarchical Models

Fitting a modified model in INLA

Description

This function takes the INLA object from the base model as input and fits weighted INLA models with weights w = 1 - δ and w = 1 + δ.

Usage

refit_inla(inla.object.base, delta)

Arguments

inla.object.base	character string, the name of the INLA object fitted in inla from the base (non-weighted, w = 1) model
delta	numeric, numerical differentiation step, the weighting factor w = 1 \pm δ, the default value is 0.01.

Details

To access the weights (w) used in INLA fit use inla.object.base\$.args\$weights. The function refit_inla modifies the original model given by inla.object.base, once with w = 1-δ and then with w = 1-δ and fits the modified models again in INLA.

Value

list composed of two INLA objects

Note

Read more details regarding the weights argument in inla.

See Also

inla

hunansona/ed4bhm documentation built on June 15, 2022, 6:42 p.m.