Skip to content

3.2-exam1-2008-strang-b

Quiz 1, 2008 (strang) · Section 3.2: Computing the Nullspace by Elimination: A = CR Source: exam1_1.pdf, pp. 199-200

Problem

4 (28 pts.) Start with this 2 by 4 matrix:

\[ A = \begin{bmatrix} 2 & 3 & 1 & -1 \\ 6 & 9 & 3 & -2 \end{bmatrix} \]

(a) Find all special solutions to \(Ax = 0\) and describe the nullspace of \(A\).

(b) Find the complete solution—meaning all solutions \((x_1, x_2, x_3, x_4)\)—to

\[ Ax = \begin{bmatrix} 2x_1 + 3x_2 + x_3 - x_4 \\ 6x_1 + 9x_2 + 3x_3 - 2x_4 \end{bmatrix} = \begin{bmatrix} 1 \\ 2 \end{bmatrix}. \]

(c) When an \(m\) by \(n\) matrix \(A\) has rank \(r = m\), the system \(Ax = b\) can be solved for which \(b\) (best answer)? How many special solutions to \(Ax = 0\)?

Solution (a) We find the special solutions by reducing \(A\). I'll go all the way to row reduced echelon form: \[ \begin{bmatrix} 2 & 3 & 1 & -1 \\ 6 & 9 & 3 & -2 \end{bmatrix} \rightsquigarrow \begin{bmatrix} 2 & 3 & 1 & -1 \\ 0 & 0 & 0 & 1 \end{bmatrix} \rightsquigarrow \begin{bmatrix} 1 & 3/2 & 1/2 & 0 \\ 0 & 0 & 0 & 1 \end{bmatrix} \] Now, in turn we set each free variable to 1 and the rest to 0, and solve \(Ux = 0\). The special solutions are \[ s_1 = \begin{bmatrix} -3/2 \\ 1 \\ 0 \\ 0 \end{bmatrix}, \qquad s_2 = \begin{bmatrix} -1/2 \\ 0 \\ 1 \\ 0 \end{bmatrix} \] The nullspace is all linear combinations of these two vectors; it will be a plane in \(\mathbb{R}^4\). (b) One way to solve for a particular solution is just to look at the set-up: our \(b\) is the negative of the 4th column, so \((0, 0, 0, -1)\) will work. We could also solve for a particular solution using an augmented matrix: \[ \begin{bmatrix} 2 & 3 & 1 & -1 & 1 \\ 6 & 9 & 3 & -2 & 2 \end{bmatrix} \rightsquigarrow \begin{bmatrix} 2 & 3 & 1 & -1 & 1 \\ 0 & 0 & 0 & 1 & -1 \end{bmatrix} \] We can now back substitute and solve. We can pick the free variables to be whatever we like; we may as well set them to be 0. Then the bottom equation gives us \(x_4 = -1\), and the top then gives \(x_1 = 0\). This is the same vector as above. The complete solution is particular solution plus the nullspace: \[ x_c = \begin{bmatrix} 0 \\ 0 \\ 0 \\ -1 \end{bmatrix} + c_1 \begin{bmatrix} -3/2 \\ 1 \\ 0 \\ 0 \end{bmatrix} + c_2 \begin{bmatrix} -1/2 \\ 0 \\ 1 \\ 0 \end{bmatrix} \] (c) When an \(m\) by \(n\) matrix has rank \(m\), the dimension of the column space (= rank) is the same as the dimension of the ambient space. Thus, every vector is in the column space; the equation \(Ax = b\) can be solved for every \(b\). The number of special solutions will be the number of free variables, \(n - r = n - m\). Thus, we find that there are \(n - m\) special solutions.

Keywords: special solutions, nullspace, complete solution, rank