em.Rd
em
is a em imputation function that returns categorical
columns results both in factor and in onehot probability vector form.
Please find the detailed documentation of em.mix
and imp.mix
in the 'mix' package. Only the modifications are explained on this page.
After the estimation of parameter pi
in em.mix
, we change it
into a tensor (multidimensional array) and
extract the probability vector from this tensor with the help of function
prob_vector_cat
.
em(df, col_cat)
Data matrix with missing values.
Categorical columns index
ximp
imputed data matrix.
ximp.disj
imputed data matrix of same type as 'ximp' for the
numeric columns.
For the categorical columns, the prediction of probability for each
category is shown in form of onehot probability vector.