kmbayes_continue: Continue sampling from existing bkmr fit
In bkmrhat: Parallel Chain Tools for Bayesian Kernel Machine Regression

Description Usage Arguments Details Value See Also Examples

Use this when you've used MCMC sampling with the kmbayes function, but you did not take enough samples and do not want to start over.

1	kmbayes_continue(fit, ...)

`fit`	output from `kmbayes`
`...`	arguments to `kmbayes_continue`

Note this does not fully start from the prior values of the MCMC chains. The kmbayes function does not allow full specification of the kernel function parameters, so this will restart the chain at the last values of all fixed effect parameters, and start the kernel r parmeters at the arithmetic mean of all r parameters from the last step in the previous chain.

a bkmrfit.continued object, which inherits from bkmrfit objects similar to kmbayes output, and which can be used to make inference using functions from the bkmr package.

kmbayes_parallel

set.seed(111)
dat <- bkmr::SimData(n = 50, M = 4)
y <- dat$y
Z <- dat$Z
X <- dat$X
## Not run: 
fitty1 = bkmr::kmbayes(y=y,Z=Z,X=X, est.h=TRUE, tier=100)
# do some diagnostics here to see if 100 iterations (default) is enough
# add 100 additional iterations (for illustration - still will not be enough)
fitty2 = kmbayes_continue(fitty1, iter=100)
cobj = as.mcmc(fitty2)
varnames(cobj)


## End(Not run)