crossprod(x) # like crossprod(x, x) but faster crossprod(x, y) x %c% y # operator form
t(x) %*% y, where
%*% is matrix
multiplication and
t is transposition.
Thus the
[i,j]th element of the result is
sum(x[,i]*y[,j]).
It is better to use
crossprod(x)
than
crossprod(x, x).
This is faster, and on some machines is more accurate.
Any computation involving
NA results in
NA.
Vectors are taken to be columns, so
crossprod(vec1,vec2) is a one by one
matrix with the element being the dot product of the two vectors.
amat <- matrix(c(19,8,11,2,18,17,15,19,10), nrow = 3)
crossprod(amat) # if amat has dimmensions of n(rows) and p(cols)
# the resultant matrix will be p by p
bmat <- c(9, 5, 14)
crossprod(amat, bmat)
amat %c% bmat