calculate_VCsquared {RCTS}

Calculates VC², to determine the stability of the found number of groups and factors over the subsamples.

Description

VC² depends on C (this is the scale parameter in PIC). When VC² is equal to zero, the found number of groups and factors are the same over the subsamples.

Usage

calculate_VCsquared(
  rcj,
  rc,
  C_candidates,
  indices_subset,
  Smax,
  limit_est_groups = 20
)

Arguments

rcj

dataframe containg the numer of groupfactors for all candidate C's and all subsamples

rc

dataframe containg the numer of common factors for all candidate C's and all subsamples

C_candidates

candidates for C (parameter in PIC)

indices_subset

all indices of the subsets

Smax

maximum allowed number of estimated groups

limit_est_groups

maximum allowed number of groups that can be estimated

Value

numeric vector with the VC2-value for each candidate C

Examples

rcj <- data.frame(X1 = rep("3_3_3", 5), X2 = rep("3_2_1", 5))
rc <- data.frame(X1 = rep(1, 5), X2 = rep(0, 5))
calculate_VCsquared(rcj, rc, 1:5, 0:1, 3)