SpatioTemporal (version 1.1.9.1)

density.mcmcSTmodel: Kernel Density Estimation for an mcmcSTmodel Object

Description

density method for class mcmcSTmodel.

Usage

# S3 method for mcmcSTmodel
density(x, BurnIn = 0, estSTmodel = NULL, ...)

Arguments

x

mcmcSTmodel object

BurnIn

Number of initial points to ignore.

estSTmodel

Either a estimateSTmodel object from estimate.STmodel or a matrix with parameter-estimates and standard deviations, such as the output from coef.estimateSTmodel. If given as a matrix, it should have columns named "par" and "sd", and rows named after the parameters.

...

Additional parameters passed to density.

Value

List containing density estimate and Gaussian densities for all model parameters.

Details

Computes kernel density estimates for the MCMC-parameters; as well as approximate Gaussian densities based on the Fischer-information.

See Also

Other mcmcSTmodel methods: MCMC.STmodel, plot.density.mcmcSTmodel, plot.mcmcSTmodel, print.mcmcSTmodel, print.summary.mcmcSTmodel, summary.mcmcSTmodel

Examples

# NOT RUN {
##load estimation results
data(est.mesa.model)
##and MCMC results instead
data(MCMC.mesa.model)

##compute density estimates for the results, and use the Gaussian approximation
##based on Fischer information as reference.
dens <- density(MCMC.mesa.model, estSTmodel=est.mesa.model)

##all the estimated densities
str(dens,1)

##or results for one paramter
dens[[1]]

##plot density functions
plot(dens)
##for a different paramter, along with Gaussian approx
plot(dens, 3, norm.col="red")

##all covariance parameters
par(mfrow=c(3,3),mar=c(4,4,2.5,.5))
for(i in 9:17){
  plot(dens, i, norm.col="red")
}
# }