Fit Besag Small Area Estimation model using aghq.

Random effects have an improper conditional autoregressive (ICAR) distribution with (generalised) precision matrix produced using the nb_to_precision function with input nb, the neighbourhood structure of sf.

Usage

besag_aghq(sf, k = 3, ii = NULL)

Arguments

sf

A simple features object with some geometry.

k

Number of quadrature points per hyperparameter dimension.

ii

The (zero-indexed) indices of the observations held-out.

Examples

besag_aghq(mw)
#> Error in .class1(object): object 'mw' not found