Skip to contents

Compute Credible Intervals for an hbmfit Object

Source: R/posterior-methods.R
posterior_interval.Rd

The posterior_interval generic is re-exported from rstantools and an S3 method is provided that dispatches on hbmfit objects. This lets users call posterior_interval(fit) on the return value of hbm just as they would on a brmsfit.

Usage

# S3 method for class 'hbmfit'
posterior_interval(object, prob = 0.95, params = NULL, ...)

Arguments

object

An hbmfit object.

prob

Coverage probability in \((0, 1)\) (default 0.95; note that rstantools::posterior_interval's own default is 0.9).

params

Optional character vector of parameter names to keep.

...

Additional arguments forwarded to posterior_draws.

Value

A matrix with two rows giving lower and upper bounds.

See also

posterior_interval

Examples

# \donttest{
library(hbsaems)
library(brms)
data("data_fhnorm")
model <- hbm(brms::bf(y ~ x1), data = data_fhnorm,
             re = ~ (1 | regency),    # area-level random effect
             chains = 4, iter = 2000, warmup = 1000,
             cores = 1, seed = 1, refresh = 0)
#> Compiling Stan program...
#> Error in .fun(model_code = .x1): Boost not found; call install.packages('BH')
posterior_interval(model, prob = 0.90)
#> Error: object 'model' not found
# }