Skip to content

3.5-exam1-2015

Exam I, 2015 · Section 3.5: Dimensions of the Four Subspaces Source: exam1_2.pdf, pp. 81, 85-86

Problem

1 (36 points) Start with the matrix

\[ A = \begin{bmatrix} 1 & -1 & 2 & 0 \\ 2 & -2 & 4 & 0 \\ 3 & -3 & 7 & 0 \end{bmatrix}. \]

(a) Find a basis for the column space \(C(A)\). (b) Find a basis for the null space \(N(A)\). (c) Find a basis for the row space \(C(A^T)\). (d) Write the complete solution to \(Ax = b\), with

\[ b = \begin{bmatrix} 1 \\ 2 \\ 4 \end{bmatrix}. \]
Solution We can notice that the second row is 2 times the first row. This means that the rows are dependent and the rank of the matrix is less than 3. In addition, we see that not all the rows are multiples of the same row, so the rank of the matrix is more than 1. Therefore it must be equal to 2. It follows that \(\dim C(A) = \dim C(A^T) = 2\), and the dimension of the nullspace is \(n - r = 4 - 2 = 2\). We can also find the dimensions by calculating \(U\): as the second row is proportional to the first one, we swap the second and the third row and have a \(PA = LU\) decomposition: \[ U = \begin{bmatrix} 1 & -1 & 2 & 0 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 0 \end{bmatrix}. \] (a) Any two independent columns of \(A\) would form a basis: columns 1 and 3, or columns 2 and 3. The default answer is to pick the pivot columns: columns 1 and 3, i.e. \[ (1, 2, 3) \quad \text{and} \quad (2, 4, 7). \] (b) The fact that the last column is zero means \((0, 0, 0, 1)\) is in the nullspace. The fact that the first column plus the second column is zero means \((1, 1, 0, 0)\) is in the nullspace. Or, we can use \(U\), and the default basis for the nullspace is \[ (1, 1, 0, 0) \quad \text{and} \quad (0, 0, 0, 1). \] (c) Similarly to (a), we can pick any two independent rows of \(A\) as the basis: row 1 and row 3, or row 2 and row 3. Or we can pick all the nonzero rows of \(U\). The most common bases are either \[ (1, -1, 2, 0) \quad \text{and} \quad (3, -3, 7, 0), \] or \((1, -1, 2, 0)\) and \((0, 0, 1, 0)\). (d) Using the row swap and the elimination on \(b\), the augmented matrix becomes \[ \left[ \begin{array}{cccc|c} 1 & -1 & 2 & 0 & 1 \\ 0 & 0 & 1 & 0 & 1 \\ 0 & 0 & 0 & 0 & 0 \end{array} \right]. \] The particular solution is \((-1, 0, 1, 0)\), and the complete solution is \[ (-1, 0, 1, 0) + a(1, 1, 0, 0) + b(0, 0, 0, 1). \]

Keywords: column space, nullspace, row space, complete solution