Skip to contents

Compute the optimal df and weighted-df using 5 spline fitting metric

Source: R/df_search.R
get_eigen_DF.Rd

Compute the optimal degree of freedom (df) and weighted degree of freedom (wdf) using 5 fitting metrics (CV: Cross-Validation, GCV: Generalised Cross-Validation, AIC: Akaike Information Criterion, BIC: Bayesian Information Criterion, AICc: Akaike Information Criterion Corrected for small sample size) over all eigenSplines generated by get_eigen_spline. The degree of freedom (df) is obtained by averaging the optimal df across each eigenSpline. The weighted degree of freedom (wdf) is obtained by weighting the optimal df in each eigenSpline by the percentage of variance explained by each eigenSpline, before summing the optimal dfs (variance sums to 100%).

Usage

get_eigen_DF(eigen)

Arguments

eigen

A list of eigenSpline parameters as generated by get_eigen_spline, containing eigen$matrix, eigen$variance, eigen$model and eigen$countTP.

Value

A list: answer$df a vector of optimum df by CV, GCV, AIC, BIC, AICc. answer$wdf a vector of weighted optimum df by CV, GCV, AIC, BIC, AICc.

See also

Graphical implementation with santaR_start_GUI

Other DFsearch: get_eigen_DFoverlay_list(), get_eigen_spline(), get_param_evolution(), plot_nbTP_histogram(), plot_param_evolution()

Examples

## 8 subjects, 8 time-points, 3 variables
inputData <- acuteInflammation$data[,1:3]
ind       <- acuteInflammation$meta$ind
time      <- acuteInflammation$meta$time
eigen     <- get_eigen_spline(inputData, ind, time, nPC=NA, scaling="scaling_UV",
                              method="nipals", verbose=TRUE, centering=TRUE, ncores=0)
#> nipals calculated PCA
#> Importance of component(s):
#>                  PC1    PC2     PC3      PC4    PC5      PC6
#> R2            0.8924 0.0848 0.01055 0.006084 0.0038 0.002362
#> Cumulative R2 0.8924 0.9772 0.98775 0.993838 0.9976 1.000000
#> total time: 0.35 secs
# nipals calculated PCA
# Importance of component(s):
#                  PC1    PC2     PC3      PC4    PC5      PC6
# R2            0.8924 0.0848 0.01055 0.006084 0.0038 0.002362
# Cumulative R2 0.8924 0.9772 0.98775 0.993838 0.9976 1.000000
get_eigen_DF(eigen)
#> $df
#>       CV      GCV      AIC      BIC     AICc 
#> 3.362581 4.255487 3.031260 2.919159 2.172547 
#> 
#> $wdf
#>       CV      GCV      AIC      BIC     AICc 
#> 2.293130 2.085212 6.675608 6.671545 4.467724 
#> 
# $df
#       CV      GCV      AIC      BIC     AICc 
# 3.362581 4.255487 3.031260 2.919159 2.172547 
# $wdf
#       CV      GCV      AIC      BIC     AICc 
# 2.293130 2.085212 6.675608 6.671545 4.467724