findMCMC_strong_corrs {runMCMCbtadjust}

findMCMC_strong_corrs

Description

finds the couples of parameters of a MCMC.list object that have at least a minCorr level of (absolute) correlation

Usage

findMCMC_strong_corrs(
  mcmcList,
  corrMethod = "pearson",
  minCorr = 0.3,
  namesToRemove = NULL
)

Arguments

mcmcList

R object of type mcmc.list that contains the MCMC output

corrMethod

character: designates the kind of correlation calculated among "pearson" (the default, for linear relationships), "spearman" (for monotone relationships) or "hoeffd" (for general associations - i.e. dependencies - between parameters)

minCorr

double, between 0 and 1: minimum level of (absolute) correlation to report.

namesToRemove

R object (can be a vector, matrix, array, list...) all components of which must be of character type: will remove parameters whose names partially match one of tghese components.

Details

In case corrMethod equals "hoeffd", the hoeffd function in the Hmisc package will be used. This can be very slow. Therefore a warning message is printed in this case.

Examples

 #\code{
# for examples with Nimble or Greta, see the Vignette.
# condition variable of whether installation is OK with Jags to avoid error durong package check
condition_jags<-TRUE
if (nchar(system.file(package='rjags'))==0) {condition_jags<-FALSE}
if (nchar(system.file(package='runjags'))==0) {condition_jags<-FALSE}
if (condition_jags)
{suppressWarnings(temp<-runjags::testjags(silent=TRUE))
 if(!(temp$JAGS.available&temp$JAGS.found&temp$JAGS.major==4)) {condition_jags<-FALSE}}

if (condition_jags) {
#generating data
set.seed(1)
y1000<-rnorm(n=1000,mean=600,sd=30)
ModelData <-list(mass = y1000,nobs = length(y1000))

#writing the Jags code as a character chain in R
modeltotransfer<-"model {

# Priors
population.mean ~ dunif(0,5000)
population.sd ~ dunif(0,100)

# Precision = 1/variance: Normal distribution parameterized by precision in Jags
population.variance <- population.sd * population.sd
precision <- 1 / population.variance

# Likelihood
for(i in 1:nobs){
  mass[i] ~ dnorm(population.mean, precision)
 }
 }"

#specifying the initial values
ModelInits <- function()
{list (population.mean = rnorm(1,600,90), population.sd = runif(1, 1, 30))}
params <- c("population.mean", "population.sd", "population.variance")
K<-3
set.seed(1)
Inits<-lapply(1:K,function(x){ModelInits()})

# running runMCMC_btadjust with MCMC_language="Jags":
set.seed(1)
out.mcmc.Coda<-runMCMC_btadjust(MCMC_language="Jags", code=modeltotransfer,
data=ModelData,
Nchains=K, params=params, inits=Inits,
niter.min=1000, niter.max=300000,
nburnin.min=100, nburnin.max=200000,
thin.min=1, thin.max=1000,
neff.min=1000, conv.max=1.05,
control=list(print.diagnostics=TRUE, neff.method="Coda"))

findMCMC_strong_corrs(out.mcmc.Coda)
}
#}