running tidyLPA after mice

[Jennifer Schauer]

Hi everyone,

since I got critical comments on using single imputation for LPA, I am now trying to do multiple imputation. Is there a way to combine tidyLPA and with( ) function?

My syntax looks like this [1] but I get an error message [mclust does not allow for missing data]. Any hints or thoughts?

Thanks a lot

Jennifer

[1] Syntax (simplified)

# mice

imp_lpa <- mice(data, m=10, maxit=5)

# lpa after mice

m_1 <- with(imp_lpa %>%
  dplyr::select(var1, var2, ...) %>%
  estimate_profiles(1:10, variances = "equal", covariances = "zero") %>%
  compare_solutions(statistics = c("LogLik", "AIC", "BIC", "SABIC", "BLRT_p", "n_min",
                                   "Entropy", "prob_min", "prob_max")))