Example

Consider a linear system $ A {\mathbf x}= {\mathbf b}$ which after the application of the Gauss-Jordan method reduces to a matrix $ [C \; \; {\mathbf d}]$ with

$\displaystyle [C \;\; {\mathbf d}] = \begin{bmatrix}1 & 0 & 2 & -1 & 0 & 0 & 2 ...
...\ 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\ 0
& 0 & 0 & 0 & 0 & 0 & 0 & 0 \end{bmatrix}.$

For this particular matrix $ [C \;\; {\mathbf d}],$ we want to see the set of solutions. We start with some observations.
Observations:
  1. The number of non-zero rows in $ C$ is $ 4.$ This number is also equal to the number of non-zero rows in $ [C \;\; {\mathbf d}].$
  2. The first non-zero entry in the non-zero rows appear in columns $ 1, 2, 5 $ and $ 6.$
  3. Thus, the respective variables $ x_1, x_2, x_5$ and $ x_6$ are the basic variables.
  4. The remaining variables, $ x_3, x_4$ and $ x_7$ are free variables.
  5. We assign arbitrary constants $ k_1, k_2$ and $ k_3$ to the free variables $ x_3, x_4$ and $ x_7,$ respectively.
Hence, we have the set of solutions as
$\displaystyle \begin{bmatrix}x_1\\ x_2\\ x_3\\ x_4\\ x_5\\ x_6\\ x_7
\end{bmatrix}$ $\displaystyle =$ $\displaystyle \begin{bmatrix}8 - 2 k_1 + k_2 - 2 k_3\\ 1 -
k_1 - 3 k_2 - 5 k_3 \\ k_1\\ k_2\\ 2 + k_3 \\ 4 -
k_3\\ k_3\end{bmatrix}$  
  $\displaystyle =$ $\displaystyle \begin{bmatrix}8\\ 1\\ 0\\
0\\ 2\\ 4\\ 0
\end{bmatrix} + k_1 \be...
...
\end{bmatrix} + k_3 \begin{bmatrix}-2\\ -5\\ 0\\ 0\\ 1\\ -1\\ 1
\end{bmatrix},$  

where $ k_1, k_2$ and $ k_3$ are arbitrary.
Let $ {\mathbf u}_0=\begin{bmatrix}8\\ 1\\ 0\\ 0\\ 2\\ 4\\ 0
\end{bmatrix}, \; {\mat...
...rix}, \; {\mathbf u}_2= \begin{bmatrix}1\\ -3\\ 0\\ 1\\ 0\\ 0\\ 0
\end{bmatrix}$ and $ {\mathbf u}_3 = \begin{bmatrix}-2\\ -5\\ 0\\ 0\\ 1\\ -1\\ 1
\end{bmatrix}.$
Then it can easily be verified that $ C {\mathbf u}_0 = {\mathbf d}, $ and for $ 1 \leq i \leq 3,\;$ $ C {\mathbf u}_i = {\mathbf 0}.$

A similar idea is used in the proof of the next theorem and is omitted. The interested readers can read the proof in Appendix 14.1.

A K Lal 2007-09-12