store_boot_def {semhelpinghands}

Store Bootstrap Estimates of User-Defined Parameters

Description

It receives a lavaan::lavaan object fitted with bootstrapping standard errors requested, computes the user-defined parameters in each bootstrap samples, and returns a lavaan::lavaan object with the estimates stored.

Usage

store_boot_def(object, force_run = FALSE)

get_boot_def(object)

Arguments

object

A lavaan object, fitted with 'se = "boot"'.

force_run

If TRUE, will skip checks and run models without checking the estimates. For internal use. Default is FALSE.

Details

lavaan::lavaan() and its wrappers, such as lavaan::sem() and lavaan::cfa(), stores the estimates of free parameters in each bootstrap sample if bootstrapping is requested. However, if a model has user-defined parameters, their values in each bootstrap sample are not stored. store_boot_def() computes the retrieves the stored bootstrap estimates and computes the values of user-defined parameters. The values are then stored in the slot external of the object, in the element shh_boot_def. The bootstrap estimates can then be used by other functions for diagnostics purposes.

get_boot_def() extracts the bootstrap estimates of user-defined parameters from a lavaan object. If none is stored, NULL is returned.

store_boot_def() is usually used with diagnostic functions such as plot_boot().

Value

store_boot_def() returns the fit object set to object, with the bootstrap values of user-defined parameters in the bootstrap samples, as a matrix, stored in the slot external of object under the name shh_boot_def.

get_boot_def() returns a matrix of the stored bootstrap estimates of user-defined parameters

Author(s)

Shu Fai Cheung https://orcid.org/0000-0002-9871-9448.

See Also

plot_boot()

Examples

library(lavaan)
set.seed(5478374)
n <- 50
x <- runif(n) - .5
m <- .40 * x + rnorm(n, 0, sqrt(1 - .40))
y <- .30 * m + rnorm(n, 0, sqrt(1 - .30))
dat <- data.frame(x = x, y = y, m = m)
model <-
'
m ~ a*x
y ~ b*m
ab := a*b
'

# Should set bootstrap to at least 2000 in real studies
fit <- sem(model, data = dat, fixed.x = FALSE,
           se = "boot",
           bootstrap = 100)
summary(fit)

# store_boot_def() is usually used with plot_boot()
# First, store the bootstrap estimates of user-defined
# parameters
fit_with_boot_def <- store_boot_def(fit)
# Second, plot the distribution of the bootstrap estimates of
# 'ab'
plot_boot(fit_with_boot_def, "ab", standardized = FALSE)