bootstrap function

Non-parametric bootstrap of StepMix estimator.

Non-parametric boostrap of StepMix estimator. Fit the estimator on X,Y then fit n_repetitions on resampled datasets. Repetition parameters are aligned with the class order of the main estimator.

## S3 method for class 'stepmix.stepmix.StepMix'
bootstrap(x, X = NULL, y = NULL, n_repetitions = 10, ...)
bootstrap(x, ...)

Arguments

• x: An object created with the fit function
• X: The X matrix or data.frame for the measurement part of the model
• y: The Y matrix or data.frame for the structural part of the model
• n_repetitions: The number of bootsrap sample
• ...: For future options. This option is actually unused.

Details

This methods returns a list with bootstrap samples (samples) and the log-likelihood (rep_stats).

Returns

A list containing bootstrap samples of the parameters.

References

Bolck, A., Croon, M., and Hagenaars, J. Estimating latent structure models with categorical variables: One-step versus three-step estimators. Political analysis, 12(1): 3-27, 2004.

Vermunt, J. K. Latent class modeling with covariates: Two improved three-step approaches. Political analysis, 18 (4):450-469, 2010.

Bakk, Z., Tekle, F. B., and Vermunt, J. K. Estimating the association between latent class membership and external variables using bias-adjusted three-step approaches. Sociological Methodology, 43(1):272-311, 2013.

Bakk, Z. and Kuha, J. Two-step estimation of models between latent classes and external variables. Psychometrika, 83(4):871-892, 2018

Author(s)

Éric Lacourse, Roxane de la Sablonnière, Charles-Édouard Giguère, Sacha Morin, Robin Legault, Félix Laliberté, Zsusza Bakk

Examples

## Not run:

if (reticulate::py_module_available("stepmix")) {
    require(stepmixr)
    model1 <- stepmix(n_components = 3, n_steps = 2, measurement = "continuous", progress_bar = 0)
    X <- iris[c(1:10, 51:60, 101:110), 1:4]
    fit1 <- fit(model1, X)
    fit1_bs <- bootstrap(fit1, X, n_repetitions = 5, progress_bar = FALSE)
}
## End(Not run)