R/Mahalanobis.R

In dismo: Species Distribution Modeling

# Author: Robert J. Hijmans
# contact: r.hijmans@gmail.com
# Date : December 2009
# Version 0.1
# Licence GPL v3

setClass('Mahalanobis',
	contains = 'DistModel',
	representation (
		cov = 'matrix' 
	),	
	prototype (	
	),
	validity = function(object)	{
		return(TRUE)
	}
)


if (!isGeneric("mahal")) {
	setGeneric("mahal", function(x, p, ...)
		standardGeneric("mahal"))
}	

setMethod('mahal', signature(x='data.frame', p='missing'), 
	function(x, p, ...) {
		for (i in ncol(x):1) {
			if (is.factor(x[,i])) {
				warning('variable "', colnames(x)[i], '" was removed because it is a factor (categorical)')
				x <- x[, -i]
			}
		}
		if (ncol(x) == 0) {	
			stop('no usable variables') 	
		}

		m <- new('Mahalanobis')
		
		x <- stats::na.omit(x)
		if (ncol(x) == 0) {	
			stop('no usable variables')
		}
		if (nrow(x) < 2) {
			stop('insufficient records')
		}
		
		m@presence <- x
		dots <- list(...)
		
		if (is.null(dots$cov)) {
			m@cov <- solve( var(x) )
		} else {
			if (!isTRUE(all.equal(dim(cov), dim(x)))) {
				stop('covariance matrix supplied does not fit the data')
			}
			if (!is.null(dots$inverted)) {
				if (! dots$inverted) {		
					cov <- solve( var(cov) )
				}
			} else {
				cov <- solve( var(cov) )			
			}
			m@cov <- cov
		}
		m
	}
)

setMethod('mahal', signature(x='matrix', p='missing'), 
	function(x, p, ...) {
		mahal(as.data.frame(x), ...)
	}
)

setMethod('mahal', signature(x='Raster', p='matrix'), 
	function(x, p, ...) {
		m <- extract(x, p)
		mahal(m, ...)
	}
)

setMethod('mahal', signature(x='Raster', p='data.frame'), 
	function(x, p, ...) {
		m <- extract(x, p)
		mahal(m, ...)
	}
)

setMethod('mahal', signature(x='Raster', p='SpatialPoints'), 
	function(x, p, ...) {
		m <- extract(x, p)
		mahal(m, ...)
	}
)