Estimate an auxiliary model based on a latent classification
by means of mixture modeling (see mx_mixture
).
The auxiliary model is treated as a multi-group model. All cases are used in all groups, but they are weighted by group-specific BCH weights as described in Bolck, Croon, & Hagenaars, 2004.
Arguments
- x
An object for which a method exists.
- model
An object that can be converted to an
OpenMx
model usingas_ram
.- data
A data.frame on which the auxiliary model can be evaluated.
- ...
further arguments to be passed to or from other methods.
References
Bolck, A., Croon, M., & Hagenaars, J. (2004). Estimating latent structure models with categorical variables: One-step versus three-step estimators. Political Analysis, 12(1), 3–27. doi:10.1093/pan/mph001
Examples
if(requireNamespace("OpenMx", quietly = TRUE)){
dat <- data.frame(x = iris$Petal.Length)
mixmod <- mx_profiles(dat,
classes = 2)
res <- BCH(mixmod, "y ~ 1", data = data.frame(y = iris$Sepal.Length))
}
#> Registered S3 method overwritten by 'OpenMx':
#> method from
#> predict.MxModel tidySEM
#> Running mix2 with 4 parameters
#> Running mix2 with 4 parameters
#> Running aux with 4 parameters