initials: Initial values for the spatio-temporal models.

In spTDyn: Spatially Varying and Spatio-Temporal Dynamic Linear Models

Initial values for the spatio-temporal models.

Description

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

Usage

initials(sig2eps=0.01, sig2eta=NULL, sig2beta=NULL, sig2delta=NULL,
   rhotp=NULL, rho=NULL, beta=NULL, phi=NULL)

Arguments

sig2eps	Initial value for the parameter \sigma^2_\epsilon.
sig2eta	Initial value for the parameter \sigma^2_\eta.
sig2beta	Initial value for the parameter \sigma^2_\beta for spatially varying model.
sig2delta	Initial value for the parameter \sigma^2_\delta for dynamic state-space model.
rhotp	Value for the parameter \rho for dynamic state-space model. For rhotp=1, \rho parameters are not sampled and fixed at value 1. For rhotp=0, \rho parameters are sampled from the full conditional distribution via MCMC with initial value 0.
rho	Initial value for the parameter \rho.
beta	Initial value for the parameter \beta.
phi	Initial value for the parameter \phi.

Note

Initial values are automatically given if the user does not provide these.

See Also

GibbsDyn, priors.

Examples

## 

initials<-initials(sig2eps=0.01, sig2eta=0.5, beta=NULL, phi=0.001)
initials

##

spTDyn documentation built on Sept. 11, 2024, 6:43 p.m.