spT.priors: Priors for the spatio-temporal models.

In jmarca/spTimer: Spatio-Temporal Bayesian Modelling

Description

This command is useful to assign the hyper-parameters of the prior distributions.

Usage

spT.priors(model="GP", inv.var.prior=Gamm(a=2,b=1),
  beta.prior=Norm(0,10^10), rho.prior=Norm(0,10^10))

Arguments

model	The spatio-temporal models, current input: "GP", "AR", and "GPP".
inv.var.prior	The hyper-parameter for the Gamma prior distribution (with mean = a/b) of the precision (inverse variance) model parameters (e.g., 1/σ2_ε, 1/σ2_η).
beta.prior	The hyper-parameter for the Normal prior distribution of the β model parameters.
rho.prior	The hyper-parameter for the Normal prior distribution of the ρ model parameter.

Note

If no prior information are given (assigned as NULL), then it use flat prior values of the corresponding distributions.
Gamm and Norm refers to Gamma and Normal distributions respectively.

See Also

spT.Gibbs, predict.spT, spT.initials.

Examples

## 

priors<-spT.priors(model="GPP",inv.var.prior=Gamm(2,1),
      beta.prior=Norm(0,10^4))
priors

##

jmarca/spTimer documentation built on May 19, 2019, 1:51 p.m.