draw_parameters: Draw plausible parameter instantiations from a given model

In mirt: Multidimensional Item Response Theory

Draw plausible parameter instantiations from a given model

Description

Draws plausible parameters from a model using parametric sampling (if the information matrix was computed) or via bootstrap sampling. Primarily for use with the DRF function.

Usage

draw_parameters(
  mod,
  draws,
  method = c("parametric", "boostrap"),
  redraws = 20,
  verbose = FALSE,
  ...
)

Arguments

mod	estimated single or multiple-group model
draws	number of draws to obtain
method	type of plausible values to obtain. Can be 'parametric', for the parametric sampling scheme which uses the estimated information matrix, or 'boostrap' to obtain values from the boot function. Default is 'parametric'
redraws	number of redraws to perform when the given parameteric sample does not satisfy the upper and lower parameter bounds. If a valid set cannot be found within this number of draws then an error will be thrown
verbose	logical; include additional information in the console?
...	additional arguments to be passed

Value

returns a draws x p matrix of plausible parameters, where each row correspeonds to a single set

Examples

set.seed(1234)
n <- 40
N <- 500

# only first 5 items as anchors
model <- 'F = 1-40
          CONSTRAINB = (1-5, a1), (1-5, d)'

a <- matrix(1, n)
d <- matrix(rnorm(n), n)
group <- c(rep('Group_1', N), rep('Group_2', N))

## -------------
# groups completely equal
dat1 <- simdata(a, d, N, itemtype = 'dich')
dat2 <- simdata(a, d, N, itemtype = 'dich')
dat <- rbind(dat1, dat2)
mod <- multipleGroup(dat, model, group=group, SE=TRUE,
                     invariance=c('free_means', 'free_var'))

param_set <- draw_parameters(mod, 100)
head(param_set)

mirt documentation built on Sept. 10, 2025, 1:07 a.m.