dalmatian (version 0.3.0)

setJAGSInits: Set initial values for dalmatian

Description

Set initial values for dalmatian

Usage

setJAGSInits(mean.model, variance.model, fixed.mean = NULL,
  fixed.variance = NULL, y = NULL, random.mean = NULL, sd.mean = NULL,
  random.variance = NULL, sd.variance = NULL)

Arguments

mean.model

Model list specifying the structure of the mean. (list)

variance.model

Model list specifyint the structure of the variance. (list)

fixed.mean

Initial values for the fixed effects of the mean. (numeric)

fixed.variance

Initial values for the fixed effects of the variance. (numeric)

y

Inital values for the true response. This should only be specified if the rounding = TRUE in the main call to dalmatian.

random.mean

Initial values for the random effects of the mean. (numeric)

sd.mean

Initial values for the standard deviation of the random effects of the mean. (numeric)

random.variance

Initial values for the random effects of the variance. (numeric

sd.variance

Initial values for the standard deviation of the random effects of the variance. (numeric)

Value

inits (list)

Details

Allows the user to set initial values for dalmatian. Any values not specified will by initialized by JAGS.

Examples

# NOT RUN {
## Load pied flycatcher data
data(pied_flycatchers_1)

## Create variables bounding the true load
pfdata$lower=ifelse(pfdata$load==0,log(.001),log(pfdata$load-.049))
pfdata$upper=log(pfdata$load+.05)

## Load output from previously run model
load(system.file("Pied_Flycatchers_1","pfresults.RData",package="dalmatian"))

## Set initial values for a new run of the same model
inits <- lapply(1:3,function(i){
  setJAGSInits(pfresults$mean.model,
               pfresults$variance.model,
               y = runif(nrow(pfdata),pfdata$lower,pfdata$upper),
               fixed.mean = tail(pfresults$coda[[i]],1)[1:4],
               fixed.variance = tail(pfresults$coda[[i]],1)[5:7],
               sd.mean = 1)
})
# }

Run the code above in your browser using DataCamp Workspace