vegdist(x, method="bray", diag=FALSE, upper=FALSE)dist and return a
distance object of the same type.euclidean
$d_{jk} = \sqrt{\sum_i (x_{ij}-x_{ik})^2}$
manhattan
$d_{jk} = \sum_i |x_{ij} - x{ik}|$
gower
$d_{jk} = \sum_i \frac{|x_{ij}-x_{ik}|}{\max_i-\min_i}$
canberra
$d_{jk}=\frac{1}{N-Z} \sum_i
\frac{|x_{ij}-x{ik}|}{x_{ij}+x_{ik}}$
bray
$\frac{\sum_i |x_{ij}-x{ik}|}{\sum_i (x_{ij}+x_{ik})}$
kulczynski
$d_{jk} = 1-0.5(\frac{\sum_i \min(x_{ij},x_{ik})}{\sum_i x_{ij}} +
\frac{\sum_i \min(x_{ij},x_{ik})}{\sum_i x_{ik}} )$
}
where $N-Z$ is the number of non-zero entries.Infamous ``double zeros'' are removed in Canberra dissimilarity.
Euclidean and Manhattan dissimilarities are not good in gradient separation without proper standardization but are still included for comparison and special needs.
Some of indices become identical or rank-order similar after some standardizations.
decostand, dist,
rankindex, isoMDSdata(varespec)
vare.dist <- vegdist(varespec)Run the code above in your browser using DataLab