x %*% y
x
and
y
.
Any computation involving a missing value will result in a missing value.
The last extent of
x
must be the same size as the first extent of
y
. Vectors are not oriented, therefore a vector of length
n
can
multiply an
n
by
n
matrix on the left or right.
If neither
x
nor
y
are matrices, then the result of x %*% y will
be a one by one matrix.
The expression
x * y
will perform element-wise multiplication,
not matrix multiplication.
freeny.ls <- lsfit(freeny.x, freeny.y) fit <- cbind(1, freeny.x) %*% freeny.ls$coef