Mantel Test for Similarity of Two Matrices
This function computes Mantel's permutation test for similarity of two matrices. It permutes the rows and columns of the two matrices randomly and calculates a Z-statistic.
mantel.test(m1, m2, nperm = 1000, graph = FALSE, ...)
- a numeric matrix giving a measure of pairwise distances, correlations, or similarities among observations.
- a second numeric matrix giving another measure of pairwise distances, correlations, or similarities among observations.
- the number of times to permute the data.
- a logical indicating whether to produce a summary graph (by default the graph is not plotted).
- further arguments to be passed to
plot()(to add a title, change the axis labels, and so on).
The function calculates a Z-statistic for the Mantel test, equal to the sum of the pairwise product of the lower triangles of the permuted matrices, for each permutation of rows and columns. It compares the permuted distribution with the Z-statistic observed for the actual data.
graph = TRUE, the functions plots the density estimate of
the permutation distribution along with the observed Z-statistic as a
... argument allows the user to give further options to
plot function: the title main be changed with
the axis labels with
xlab =, and
ylab =, and so on.
z.stat the Z-statistic (sum of rows*columns of lower triangle) of the data matrices. p P-value (quantile of the observed Z-statistic in the permutation distribution).
Mantel, N. (1967) The detection of disease clustering and a generalized regression approach. Cancer Research, 27, 209--220.
Manly, B. F. J. (1986) Multivariate statistical methods: a primer. London: Chapman & Hall.
q1 <- matrix(runif(36), nrow = 6) q2 <- matrix(runif(36), nrow = 6) mantel.test(q1, q2, graph = TRUE, main = "Mantel test: a random example with 6 X 6 matrices", xlab = "z-statistic", ylab = "Density", sub = "The vertical line shows the observed z-statistic")