Skip to contents

Sample a model again with changed sampling arguments

Source: R/methods-draws.R
update.frmtmb_draws.Rd

brms's update() of a fitted model, for the arguments that change how the model is sampled and not the model itself: sample_prior, prior, and the sampler's options (chains, iter, seed, control, ...). It calls frm_sample() again on the same model with the arguments of the first call, the ones given here replacing them. A new prior merges as in brms: it takes the slots it names, and the first call's other specifications stay. brms's update(fit, sample_prior = "yes") is the common use: it adds prior draws to a model sampled without them, for prior_draws() and the evidence ratios of frmtmb::hypothesis().

Usage

# S3 method for class 'frmtmb_draws'
update(object, formula., newdata = NULL, recompile = NULL, ...)

Arguments

object

A frmtmb_draws from frm_sample().

formula., newdata

brms's arguments for a new model; refused.

recompile

brms's argument for a new Stan program; frmtmb compiles none, and anything but NULL is refused.

...

Arguments of frm_sample() to change.

Value

A new frmtmb_draws.

Details

A new formula or new data changes the model, which the draws do not carry in a form that can be revised: update() the fit with them and call frm_sample() on the result.

Examples

# \donttest{
if (requireNamespace("tmbstan", quietly = TRUE) &&
    requireNamespace("rstan", quietly = TRUE) &&
    !frmtmb.sample:::tmbstan_build_broken()) {
  set.seed(1)
  dd <- data.frame(x = rnorm(40))
  dd$y <- rnorm(40, 1 + 0.5 * dd$x, 1)
  ds <- frm_sample(bf(y ~ x), family = gaussian(), data = dd,
                   prior = set_prior("normal(0, 1)", class = "b"),
                   chains = 1, iter = 400, seed = 1, refresh = 0)
  ds2 <- update(ds, sample_prior = "yes")
  names(prior_draws(ds2))
}
#> frm_sample(): default priors (brms 2.23 defaults; prior = "flat" puts brms's (flat) in every slot)
#>   Intercept          student_t(3, 1.1, 2.5)
#>   sigma              student_t(3, 0, 2.5)  [natural scale]
#>   b                  (flat), as brms leaves slopes
#> Warning: Bulk Effective Samples Size (ESS) is too low, indicating posterior means and medians may be unreliable.
#> Running the chains for more iterations may help. See
#> https://mc-stan.org/misc/warnings.html#bulk-ess
#> frm_sample(): default priors (brms 2.23 defaults; prior = "flat" puts brms's (flat) in every slot)
#>   Intercept          student_t(3, 1.1, 2.5)
#>   sigma              student_t(3, 0, 2.5)  [natural scale]
#>   b                  (flat), as brms leaves slopes
#> Warning: Bulk Effective Samples Size (ESS) is too low, indicating posterior means and medians may be unreliable.
#> Running the chains for more iterations may help. See
#> https://mc-stan.org/misc/warnings.html#bulk-ess
#> [1] "Intercept" "b"         "sigma"    
# }