mirt (version 1.41)

draw_parameters: 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,
  ...
)

Value

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

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

Examples

Run this code

if (FALSE) {
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)
}

Run the code above in your browser using DataLab