Skip to content

3.2-exam1-2013

Quiz 1, 2013 · Section 3.2: Computing the Nullspace by Elimination: A = CR Source: exam1_2.pdf, pp. 34, 55-56

Problem

2 (40 pts.) \(A\) is a 2 by 4 matrix with exactly two special solutions to \(Ax = 0\):

\[ x = s_1 = \begin{bmatrix} 3 \\ 1 \\ 0 \\ 0 \end{bmatrix} \quad x = s_2 = \begin{bmatrix} 6 \\ 0 \\ 2 \\ 1 \end{bmatrix}. \]

(a) Find the reduced row echelon form \(R\) of \(A\). (b) What is the column space of \(A\)? (c) What is the complete solution to \(Rx = \begin{bmatrix} 3 \\ 6 \end{bmatrix}\)? (d) Find a combination of columns 2, 3, 4 that equals the zero vector. (Not OK to use \(0(\text{col } 2) + 0(\text{col } 3) + 0(\text{col } 4) = 0\). The problem is to show that these 3 columns are dependent.)

Solution (a) Any row of \(A\) is orthogonal to the two special solutions given in the problem. That is, any row \[ r = \begin{bmatrix} x \\ y \\ z \\ w \end{bmatrix} \] satisfies \(r \cdot s_1 = r \cdot s_2 = 0\). This is just a system of two linear equations, so we need to solve the equation \[ \begin{bmatrix} 3 & 1 & 0 & 0 \\ 6 & 0 & 2 & 1 \end{bmatrix} r = 0, \] whose complete solution is given by \[ c \begin{bmatrix} 1 \\ -3 \\ 0 \\ -6 \end{bmatrix} + d \begin{bmatrix} 0 \\ 0 \\ -1 \\ 2 \end{bmatrix}, \] from which we get the reduced row echelon form of \(A\): \[ R = \begin{bmatrix} 1 & -3 & 0 & -6 \\ 0 & 0 & 1 & -2 \end{bmatrix}. \] (b) \(R\) has two pivots, and therefore \(A\) has two pivots and \(\text{rank}(A) = 2\). Two independent columns in \(R^2\) span \(R^2\), so \[ C(A) = R^2. \] (c) The free variables are \(x_2, x_4\), so the particular solution is \[ x_p = \begin{bmatrix} 3 \\ 0 \\ 6 \\ 0 \end{bmatrix}. \] The complete solution is \[ x_c = x_p + c_1 s_1 + c_2 s_2 = \begin{bmatrix} 3 \\ 0 \\ 6 \\ 0 \end{bmatrix} + c_1 \begin{bmatrix} 3 \\ 1 \\ 0 \\ 0 \end{bmatrix} + c_2 \begin{bmatrix} 6 \\ 0 \\ 2 \\ 1 \end{bmatrix}. \] (d) One can find the coefficients by inspection, or by combining the special solutions: \[ 2s_1 - s_2 = \begin{bmatrix} 0 \\ 2 \\ -2 \\ -1 \end{bmatrix}. \] The vector \(2s_1 - s_2\) is in the nullspace, so in terms of the columns of \(R\): \[ 2 \cdot \begin{bmatrix} -3 \\ 0 \end{bmatrix} - 2 \cdot \begin{bmatrix} 0 \\ 1 \end{bmatrix} - \begin{bmatrix} -6 \\ -2 \end{bmatrix} = 0, \] and in terms of the columns of \(A\) (which have the same dependence relations): \[ 2(\text{col } 2) - 2(\text{col } 3) - 1(\text{col } 4) = 0. \]

Keywords: special solutions, rref, complete solution, linear dependence