




The transformation A takes the jth column basis vector into the jth column
of the matrix A; which is a linear combination of the basis vectors,
with coefficient A_{kj} for the basis vector indexed by k.
B takes the kth basis vector to the kth column of B, and takes this linear combination
of basis vectors to the same linear combination of the columns of B.
Thus the entry (BA)_{ij} consists of the sum of the B_{ik},
each weighted by A_{kj}: