Learn R Programming

ade4 (version 1.7-22)

yanomama: Distance Matrices

Description

This data set gives 3 matrices about geographical, genetic and anthropometric distances.

Usage

data(yanomama)

Arguments

Format

yanomama is a list of 3 components:

geo

is a matrix of 19-19 geographical distances

gen

is a matrix of 19-19 SFA (genetic) distances

ant

is a matrix of 19-19 anthropometric distances

References

Table 7.2 Distance matrices for 19 villages of Yanomama Indians. All distances are as given by Spielman (1973), multiplied by 100 for convenience in: Manly, B.F.J. (1991) Randomization and Monte Carlo methods in biology Chapman and Hall, London, 1--281.

Examples

Run this code
    data(yanomama)
    gen <- quasieuclid(as.dist(yanomama$gen)) # depends of mva
    ant <- quasieuclid(as.dist(yanomama$ant)) # depends of mva
    par(mfrow = c(2,2))
    plot(gen, ant)
    t1 <- mantel.randtest(gen, ant, 99);
    plot(t1, main = "gen-ant-mantel") ; print(t1)
    t1 <- procuste.rtest(pcoscaled(gen), pcoscaled(ant), 99)
    plot(t1, main = "gen-ant-procuste") ; print(t1)
    t1 <- RV.rtest(pcoscaled(gen), pcoscaled(ant), 99)
    plot(t1, main = "gen-ant-RV") ; print(t1)

Run the code above in your browser using DataLab