summary.miclust {miclust}

Summarizes the results.

Description

Performs a within-cluster descriptive analysis of the variables after the clustering process performed by the function miclust.

Usage

## S3 method for class 'miclust'
summary(object, k = NULL, quantilevars = NULL, ...)

Arguments

object

object of class miclust obtained with the function miclust.

k

number of clusters. The default value is the optimal number of clusters obtained by miclust.

quantilevars

numeric. If a variable selection procedure was used, the cut-off percentile in order to decide the number of selected variables in the variable reduction procedure by decreasing order of presence along the imputations results. The default value is quantilevars = 0.5, i.e., the number of selected variables is the median number of selected variables along the imputations.

...

further arguments for the plot function.

Value

An object with classes c("list", "summary.miclust") including the following items:

allocationprobabilities

if imputations were analyzed, descriptive summary of the probability of cluster assignment.

classmatrix

if imputations were analyzed, the individual probabilities of cluster assignment.

cluster

if imputations were analyzed, the final individual cluster assignment.

clusterssize

if imputations were analyzed, size of the imputed cluster and between-imputations summary of the cluster size.

clustervector

if a single data set (raw data set) has been clustered, a vector containing the individuals cluster assignments.

clustervectors

if imputed data sets have been clustered, the individual cluster assignment in each imputation.

completecasesperc

if a single data set (raw data set) has been clustered, the percentage of complete cases in the data set.

k

number of clusters.

kappas

if imputations were analyzed, the Cohen's kappa values after comparing the cluster vector in the first imputation with the cluster vector in each of the remaining imputations.

kappadistribution

a summary of kappas.

m

number of imputations used in the descriptive analysis which is the total number of imputations provided.

quantilevars

if variable selection was performed, the input value of quantilevars.

search

search algorithm for the selection variable procedure.

selectedvariables

if variable selection was performed, the selected variables obtained considering quantilevars.

selectedvarspresence

if imputations were analyzed and variable selection was performed, the presence of the selected variables along imputations.

summarybycluster

within-cluster descriptive analysis of the selected variables.

usedimp

indicator of imputations used in the clustering procedure.

See Also

miclust, plot.miclust.

Examples

### see examples in miclust.