Utility functions for the PC prior for correlation in AR(1)

Functions to evaluate, sample, compute quantiles and percentiles of the PC prior for the correlation in the Gaussian AR(1) model where the base-model is zero correlation.

inla.pc.rcor0(n, u, alpha, lambda)

inla.pc.dcor0(cor, u, alpha, lambda, log = FALSE)

inla.pc.qcor0(p, u, alpha, lambda)

inla.pc.pcor0(q, u, alpha, lambda)

Arguments

n

Number of observations

u

The upper limit (see Details)

alpha

The probability going above the upper limit (see Details)

lambda

The rate parameter (see Details)

cor

Vector of correlations

log

Logical. Return the density in natural or log-scale.

p

Vector of probabilities

q

Vector of quantiles

Value

inla.pc.dcor0 gives the density, inla.pc.pcor0 gives the distribution function, inla.pc.qcor0 gives the quantile function, and inla.pc.rcor0 generates random deviates.

Details

The statement Prob(|cor| > u) = alpha is used to determine lambda unless lambda is given. Either lambda must be given, or u AND alpha. The density is symmetric around zero.

See also

inla.doc("pc.rho0")

Author

Havard Rue hrue@r-inla.org

Examples

 cor = inla.pc.rcor0(100,  lambda = 1)
 d = inla.pc.dcor0(cor, lambda = 1)
 cor = inla.pc.qcor0(c(0.3, 0.7), u = 0.5, alpha=0.01)
 inla.pc.pcor0(cor, u = 0.5, alpha=0.01)
#> [1] 0.3 0.7