eigenmap: Spatial Eigenvector Maps

In guenardg/codep: Multiscale Codependence Analysis

Spatial Eigenvector Maps

Description

Function to calculate spatial eigenvector maps of a set of locations in a space with an arbitrary number of dimension.

Usage

eigenmap(
  x,
  alt.coord = NA,
  weighting = wf.sqrd,
  boundaries,
  wpar,
  tol = .Machine$double.eps^0.5
)

eigenmap.score(emap, target)

Arguments

x	A set of coordinates defined in one (numeric vector) or many (a coordinate x dimension matrix) dimensions or, alternatively, a distance matrix provided by dist.
alt.coord	Coordinates to be used when a distance matrix is provided as x. Used for plotting purposes.
weighting	The function to obtain the edge weighting matrix (see details).
boundaries	When required by argument weighting, a two-element numeric vector containing the lower and upper threshold values used to obtain the connectivity matrix (see weighting-functions).
wpar	Shape parameter for argument weignting (optional).
tol	The smallest absolute eigenvalue for a spatial eigenfunctions to be considered as a suitable predictor. Default: .Machine$double.eps^0.5 (a machine-dependent value).
emap	An eigenmap-class object.
target	A (generally rectangular) distance matrix between a set of target locations for which spatially-explicit predictions are being made (rows), and the reference locations given to function eigenmap (columns). See example 2.

Details

When function eigenmap is given coordinates as its argument x, they are treated as Cartesian coordinates and the distances between them are assumed to be Euclidean. Otherwise (e.g., when geodesic distances are used), distances have to be provided as the argument x and plotting coordinates have to be supplied as argument alt.coord.

The weighting function (see weighting-functions) must have the distances as its first argument, optionally an argument named boundaries giving the boundaries within which locations are regarded as neighbours and/or an argument wpar containing any other weighting function parameters.

Default values for argument boundaries are 0 for the minimum value and NA for the maximum. For weighting functions with an argument bounraries, The upper value NA indicates the function to take the minimum value that allow every locations to form a single cluster following single linkage clustering as a maximum value (obtained internally from a call to hclust.

Functions

• eigenmap(): Main function for generating an eigenmap-class object from Cartesian coordinates or pairwise distances.

• eigenmap.score(): Generate scores for arbitrary locations within the scope of an existing eigenvector map.

Author(s)

Guillaume Guenard and Pierre Legendre, Bertrand Pages Maintainer: Guillaume Guenard <[e-mail]>

References

Borcard, D. and Legendre, P. 2002. All-scale spatial analysis of ecological data by means of principal coordinates of neighbour matrices. Ecol. Model. 153: 51-68

Dray, S.; Legendre, P. and Peres-Neto, P. 2006. Spatial modelling: a comprehensive framework for principal coordinate analysis of neighbor matrices (PCNM). Ecol. Modelling 196: 483-493

Legendre, P. and Legendre, L. 2012. Numerical Ecology, 3rd English edition. Elsevier Science B.V., Amsterdam, The Netherlands.

Examples

### Example 1: A linear transect.
data(salmon)

## A warning is issued when no boundaries are provided for a function that
## requires them.
## Example:
map <- eigenmap(x = salmon[,"Position"], weighting = wf.binary)
map
## plot(map)

## In the following examples, boundaries are provided; they are needed by the
## functions.
map <- eigenmap(x = salmon[,"Position"], weighting = wf.binary,
                boundaries = c(0,20))
map
## plot(map)

map <- eigenmap(x = salmon[,"Position"], weighting = wf.Drayf1,
                boundaries = c(0,20))
map
## plot(map)

map <- eigenmap(x = salmon[,"Position"], weighting = wf.Drayf2,
                boundaries = c(0,20))
map
## plot(map)

map <- eigenmap(x = salmon[,"Position"], weighting = wf.Drayf3,
                boundaries = c(0,20), wpar = 2)
map
## plot(map)

map <- eigenmap(x = salmon[,"Position"], weighting = wf.PCNM,
                boundaries = c(0,20))
map
## plot(map)

map <- eigenmap(x = salmon[,"Position"], weighting = wf.sqrd)
map
## plot(map)

map <- eigenmap(x = salmon[,"Position"], weighting = wf.RBF, wpar = 0.001)
map
## plot(map)

### Example 2: Using predictor scores

smpl <- c(4,7,10,14,34,56,61,64)  # A sample to be discarded
map <- eigenmap(x = salmon[-smpl,"Position"], weighting = wf.sqrd)
scr <- eigenmap.score(
         map, target = as.matrix(dist(salmon[,"Position"]))[,-smpl]
       )
## Scores of sampling points are the eigenvectors
scr[smpl,]

wh <- 5L   # You can try with other vectors.
plot(map$U[,wh] ~ salmon[-smpl,"Position"], ylab = expression(U[5]),
     xlab = "Position along transect")
points(y = scr[smpl,wh], x = salmon[smpl,"Position"], pch = 21L,
       bg = "black")

map <- eigenmap(x = salmon[-smpl,"Position"], weighting = wf.binary,
                boundaries = c(0,20))
scr <- eigenmap.score(
         map, target = as.matrix(dist(salmon[,"Position"]))[smpl,-smpl])

## Plot the 8 prediction sites along particular eigenvectors, here
## eigenvector #1:
wh <- 1L   # One could try the other vectors.
plot(map$U[,wh] ~ salmon[-smpl,"Position"], ylab = expression(U[1L]),
     xlab = "Position along transect (m)")
points(y = scr[,wh], x = salmon[smpl,"Position"], pch=21L, bg = "black")

map <- eigenmap(x = salmon[-smpl,"Position"], weighting = wf.PCNM,
                boundaries = c(0,100))
scr <- eigenmap.score(
         map, target = as.matrix(dist(salmon[,"Position"]))[smpl,-smpl]
       )

wh <- 1L   # You can try with other vectors.
plot(map$U[,wh] ~ salmon[-smpl,"Position"], ylab = expression(U[1]),
     xlab = "Position along transect (m)")
points(y = scr[,wh], x = salmon[smpl,"Position"], pch = 21L, bg = "black")

### Example 3: A unevenly sampled surface.

data(mite)

## Example using the principal coordinates of the square root of the
## (Euclidean) distances:
map <- eigenmap(x = as.matrix(mite.geo), weighting = wf.sqrd)
map
## plot(map)

## Example using the radial basis functions (RBF):
map <- eigenmap(x = as.matrix(mite.geo), weighting = wf.RBF)
map
## plot(map)

guenardg/codep documentation built on April 16, 2024, 9:01 p.m.