A fixed teaching simulation with missing weights. Fits the same linear model to each completed dataset and pools model estimates, not datasets. It does not impute uploaded records. Diagnostic plots do not establish that the missingness assumptions hold in a real study.
Value
A list with complete and incomplete synthetic data, the mids
object, completed datasets, pooled model estimates and a comparison of
the complete-data, complete-case, mean-imputed and pooled slopes.
Examples
# \donttest{
demo <- cata_mice_demo(m = 3, maxit = 3)
demo$comparison
#> # A tibble: 4 × 5
#> method estimate lower upper records
#> <chr> <dbl> <dbl> <dbl> <int>
#> 1 Complete synthetic data 0.108 0.0971 0.119 120
#> 2 Complete cases 0.104 0.0912 0.116 84
#> 3 Single mean imputation 0.0750 0.0629 0.0872 120
#> 4 Multiple imputation (pooled) 0.102 0.0853 0.118 120
# }
