A function to add two vectors or matrices

Description

It is sometimes convenient to add two vectors or matrices in an operation analogous to matrix multiplication. For matrices nXm and mYp, the matrix sum of the i,jth element of nSp = sum(over m) of iXm + mYj.

Usage

x %+% y 

Arguments

Details

Used in such problems as Thurstonian scaling. Although not technically matrix addition, as pointed out by Krus, there are many applications where the sum or difference of two vectors or matrices is a useful operation. This can be done, of course, through

Value

a n by p matix of sums

Author(s)

William Revelle

References

Krus, D. J. (2001) Matrix addition. Journal of Visual Statistics, 1, (February, 2001).

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--    or do  help(data=index)  for the standard data sets.
x <- seq(1,4)
z <- x %+% -t(x)
x
z
x <- matrix(seq(1,6),ncol=2)
y <- matrix(seq(1,10),nrow=2)
z <- x %+% y
x
y
z
## The function is currently defined as
"%+%" <- function(x,y) { 
 if(!is.matrix(x)) {
        if(is.vector(x)) {x <- as.matrix(x)} else stop("x must be either a vector or a matrix")}
if(!is.matrix(y)) {
        if(is.vector(y)) {y <- as.matrix(y)} else stop("y must be either a vector or a matrix")}
n.x <- dim(x)[1]
n.y <- dim(y)[2]
n.k <- dim(x)[2]
if (n.k != dim(y)[1]) {warning("Matrices should be comparable")}
#first find sum vectors
x <- rowSums(x)
y <- colSums(y)
one <- as.vector(rep(1,n.y))  #to duplicate x n.y times
one.y <- as.vector(rep(1,n.x)) #to duplicate y n.x times
xy  <- x 
  return(xy) }