get_final_estimation: Function that returns the final clustering, based on the...

In RCTS: Clustering Time Series While Resisting Outliers

Function that returns the final clustering, based on the estimated number of groups and common and group specific factors.

Description

Function that returns the final clustering, based on the estimated number of groups and common and group specific factors.

Usage

get_final_estimation(df, opt_groups, k, kg, type, limit_est_groups = 20)

Arguments

df	input dataframe (this will be df_results_full)
opt_groups	the optimal number of groups
k	the optimal number of common factors
kg	vector with the optimal number of group specific factors
type	defines which estimation to return: options are "clustering", "beta", "fg" (group specific factors), "lg" (loadings corresponding to fg), "f" (common factors), "l" (loadings corresponding to f),
limit_est_groups	maximum allowed number of groups that can be estimated

Value

This function returns the estimations of the chosen configuration. If type is "clustering" it returns a numeric vector with the estimated group membership for all time series. If type is "beta", "lg" the function returns a data.frame. If type is "f" or "l" the function also returns a data.frame. If no common factors were estimated in the optimized configuration, then NA is returned. If type is "fg" the function returns a list.

Examples

get_final_estimation(df_results_example, 3, 0, c(3, 3, 3), "clustering")
get_final_estimation(df_results_example, 3, 0, c(3, 3, 3), "beta")
get_final_estimation(df_results_example, 3, 0, c(3, 3, 3), "fg")
get_final_estimation(df_results_example, 3, 0, c(3, 3, 3), "lg")

RCTS documentation built on May 31, 2023, 9:20 p.m.