Rao's diversity coefficient also called quadratic entropy
divc.Rd
Calculates Rao's diversity coefficient within samples.
Arguments
- df
a data frame with elements as rows, samples as columns, and abundance, presence-absence or frequencies as entries
- dis
an object of class
dist
containing distances or dissimilarities among elements. Ifdis
is NULL, Gini-Simpson index is performed.- scale
a logical value indicating whether or not the diversity coefficient should be scaled by its maximal value over all frequency distributions.
Value
Returns a data frame with samples as rows and the diversity coefficient within samples as columns
References
Rao, C.R. (1982) Diversity and dissimilarity coefficients: a unified approach. Theoretical Population Biology, 21, 24–43.
Gini, C. (1912) Variabilità e mutabilità. Universite di Cagliari III, Parte II.
Simpson, E.H. (1949) Measurement of diversity. Nature, 163, 688.
Champely, S. and Chessel, D. (2002) Measuring biological diversity using Euclidean metrics. Environmental and Ecological Statistics, 9, 167–177.
Author
Sandrine Pavoine pavoine@mnhn.fr
Examples
data(ecomor)
dtaxo <- dist.taxo(ecomor$taxo)
divc(ecomor$habitat, dtaxo)
#> diversity
#> Bu1 2.754458
#> Bu2 2.967895
#> Bu3 3.107200
#> Bu4 3.137755
#> Ca1 2.500000
#> Ca2 3.230769
#> Ca3 3.272727
#> Ca4 3.283203
#> Ch1 2.333333
#> Ch2 2.595041
#> Ch3 3.246528
#> Ch4 3.135734
#> Pr1 2.840000
#> Pr2 2.816327
#> Pr3 3.055556
#> Pr4 3.028355
data(humDNAm)
divc(humDNAm$samples, sqrt(humDNAm$distances))
#> diversity
#> oriental 0.43100189
#> tharu 0.48255042
#> wolof 0.65884298
#> peul 0.55952920
#> pima 0.06198035
#> maya 0.17092768
#> finnish 0.33280992
#> sicilian 0.61913580
#> israelij 0.67061144
#> israelia 0.64036818