simba (version 0.3-1)

ads.ternaries: Artificial data-set for studying the mathematical behavior of asymmetric similarity coefficients

Description

Artificial data-set as utilized in Koleff et al. 2003, and Jurasinski 2007 to study the mathematical behavior of asymmetrical similarity indices. The corresponding values of all indices computable with sim are contained. See example!

Usage

data(ads.ternaries)

encoding

UTF-8

format

• ads.ternaries

{data.frame with the three matching components of asymmetric binary similarity measures (a, b, c) with all possible combinations of these components derived from a virtual data-set with 100 variables (species). These are the first three columns. The preceding columns contain the values of the similarity coefficients computable with sim according to the three matching components. This information can be used to study the mathematical behavior of the indices. See example}

source

Jurasinski, G. (2007) Spatio-Temporal Patterns of Biodiversity and their Drivers - Method Development and a Case Study from Northeastern Morocco. PhD-Thesis, Department of Biogeography, University of Bayreuth Koleff, P., Gaston, K. J. & Lennon, J. J. (2003) Measuring beta diversity for presence-absence data. Journal of Animal Ecology 72: 367-382.

Examples

data(ads.ternaries)
library(plotrix)
##take any index you want to study, see the help for sim() for available 
##asymmetric indices or the names of the data.frame:
names(ads.ternaries)

##make a tmp from the index you want to study (we perform a transformation 
##to obtain values between 0 and 1). if you want another index, 
##just change the name in the next line:
tmp <- ads.ternaries$mountford
tmp <- (tmp-min(tmp))/max(tmp)
triax.plot(ads.ternaries[,c(2,3,1)], main="mountford", 
col.symbols=grey(seq(0.1,1,0.1))[floor((tmp*100)/5)+1], pch=16)

##don't wonder: mountford is strange, just try another one:
##this time with rainbow-colors
tmp <- ads.ternaries$soerensen
tmp <- (tmp-min(tmp))/max(tmp)
triax.plot(ads.ternaries[,c(2,3,1)], main="sørensen", 
col.symbols=rainbow(10)[floor((tmp*100)/10)+1], pch=16)

##and an interesting shape: routledge in greyscale...
tmp <- ads.ternaries$routledge
tmp <- (tmp-min(tmp))/max(tmp)
triax.plot(ads.ternaries[,c(2,3,1)], main="routledge", 
col.symbols=grey(seq(0.1,1,0.1))[floor((tmp*100)/5)+1], pch=16)