Suppose that \mathcal{S} = \{u_{1}, u_{2}, . . . , u_{n}\}[ is a set of vectors from ℜ^m. Prove that \mathcal{S} is linearly independent if and only if the set
\mathcal{S}^{′}=\left\{u_{1},\sum\limits_{i=1}^{2}u_{i}, \sum\limits_{i=1}^{3}u_{i}, . . . ,\sum\limits_{i=1}^{n}u_{i}\right\}
is linearly independent.
If A_{m×n} is the matrix containing the u_{i}’s as columns, and if
Q_{n×n} =\begin{pmatrix}1 &1& · · ·& 1\\0 &1& · · ·& 1\\\vdots&\vdots&\ddots&\vdots\\0 &0& · · · &1\end{pmatrix},
then the columns of B = AQ are the vectors in \mathcal{S}^{′}. Clearly, Q is nonsingular so that A \overset{col}{\sim} B, and thus rank (A) = rank (B). The desired result now follows from (4.3.3).
rank (A) = n. (4.3.3)