man/examples/addhealth.R

In slca: Structural Modeling for Multiple Latent Class Variables

library(magrittr)
data <- addhealth[1:200,]
lta5 <- slca(
   DEP1(5) ~ S1w1 + S2w1 + S3w1 + S4w1 + D1w1 + D2w1 + F1w1 + F2w1,
   DEP2(5) ~ S1w2 + S2w2 + S3w2 + S4w2 + D1w2 + D2w2 + F1w2 + F2w2,
   DEP1 ~ DEP2
) %>% estimate(data, control = list(em.tol = 1e-6))
lta5inv <- slca(
   DEP1(5) ~ S1w1 + S2w1 + S3w1 + S4w1 + D1w1 + D2w1 + F1w1 + F2w1,
   DEP2(5) ~ S1w2 + S2w2 + S3w2 + S4w2 + D1w2 + D2w2 + F1w2 + F2w2,
   DEP1 ~ DEP2,
   constraints = c("DEP1", "DEP2")
) %>% estimate(data, control = list(em.tol = 1e-6))

compare(lta5inv, lta5, test = "chisq")
lta5inv %>% param()