bri.adapt.prior: Construct a random walk prior for adaptive smoothing

In julianfaraway/brinla: Bayesian Regression with INLA

Construct a random walk prior for adaptive smoothing

Description

Construct a random walk prior for adaptive smoothing

Usage

bri.adapt.prior(x, degree = 3, nknot = 5, theta.prec = 0.01,
  type = c("indpt", "spde"))

Arguments

x	the predictor vector
degree	the degree of B-spline basis used for the smoothing function (default is 3)
nknot	the number of knots for the B-spline basis (default is 5)
theta.prec	the fixed precision of the Gaussian prior on the smoothing function (default value is 0.01)
type	the type of the Gaussian prior on the smoothing function: 'indpt' independent Gaussian priors; 'spde' SPDE prior

Value

an object for the 'model' option used in inla()

julianfaraway/brinla documentation built on April 6, 2023, 2:33 p.m.