CoDA Atypicality Index

Calculates the atypicality index (ranges between 0 and 1, 1 being more atypical/different) for all the observations of a composition or compares one sample to the given composition, using Compositional Data Analysis (CoDA) principles.

Usage

CoDA_Atyp_Idx(comp, sample = NULL, thres = 0.95)

Arguments

comp

A matrix or data frame of observations for a given composition. Entries must be non-zero and positive.

sample

A one sample vector to compare against the given composition.

thres

Threshold to classify observation as atypical, from 0 to 1 (Default is 0.95)

Value

A matrix with the atypicality index or indices.

References

Aitchison, J. (1986). The statistical analysis of compositional data. Chapman and Hall.

See also

Other compositional data analysis: CoDA_2Group_H1(), CoDA_Univariate()

Examples

data("Hongite", package = 'compositions')
samp = c(44, 20.4, 13.9, 9.1, 12.6)
CoDA_Atyp_Idx(Hongite)
#> New names:
#> • `` -> `...1`
#> # A tibble: 25 × 2
#>       idx Atypical
#>     <dbl> <chr>   
#>  1 0.712  No      
#>  2 0.0171 No      
#>  3 0.199  No      
#>  4 0.0318 No      
#>  5 0.709  No      
#>  6 0.828  No      
#>  7 0.882  No      
#>  8 0.969  Yes     
#>  9 0.1    No      
#> 10 0.987  Yes     
#> # ℹ 15 more rows
CoDA_Atyp_Idx(Hongite, samp)
#> New names:
#> • `` -> `...1`
#> # A tibble: 1 × 2
#>     idx Atypical
#>   <dbl> <chr>   
#> 1 0.998 Yes