Splitting the initial dataset into kstar clusters

Splitting the initial dataset into kstar clusters by using the parameter kstar determined in the module mod_kstar.

Usage

mod_calc_kamila(
  PARAM_KAMILA,
  CONT_DF,
  CATEG_DF,
  FULL_CONT_DF,
  FULL_CATEG_DF,
  KM_RES,
  list = NULL
)

PARAM_KAMILA: data frame with all needed parameters for the Kamila method, from which the following parameters are used: - numinit: The number of initializations used. - maxiter: The maximum number of iterations in each run. - param_kstar: Best number of clusters estimated in the module mod_kstar.
CONT_DF: Subset of the pension register containing all the continuous variables except for the outcome variables aadr and monthly_pension.
CATEG_DF: Subset of the pension register containing all categorical variables as factors except for the nominal variables marital_stat and benef_type.
FULL_CONT_DF: Data frame containing the continuous variables used for the estimation plus the outcome variables aadr and monthly_pension.
FULL_CATEG_DF: Data frame containing the categorical variables used for the estimation plus the nominal variables marital_stat and benef_type.
KM_RES: Tibble contains the kamila results of the training set.
list: List of input data frames.

a tidylist containing the following tidy data frames:

PLOTDATKAM Data frame containing the clusters factor and the other variables.
KM_RES_FINAL Data frame containing the resulting parameters of the clustering.
CONTVARS Data frame containing the continuous standardised variables.
FULL_CONT_DF Data frame containing the continuous variables used for the estimation.
FULL_CATEG_DF Data frame containing the categorical variables used for the estimation.