Skip to content

4.4-exam2-2000-spring

Exam 2, 2000 (spring) · Section 4.4: Orthonormal Bases and Gram-Schmidt Source: exam2_1.pdf, pp. 48, 52

Problem

1 (30 pts) Let

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

(a) Find orthonormal vectors \(q_1, q_2\), and \(q_3\) so that \(q_1\) and \(q_2\) form a basis for the column space of \(A\).

(b) Which of the four fundamental subspaces contains \(q_3\)?

(c) Find the projection matrix \(P\) projecting onto the left nullspace (not the column space!) of \(A\).

(d) Find the least squares solution to \(Ax = (1, 2, 7)\).

Note: You must show your work to receive credit for this problem.

Solution 1a. \[ q_1 = \frac{1}{3} \begin{bmatrix} 1 \\ 2 \\ -2 \end{bmatrix}, \qquad q_2 = \frac{1}{3} \begin{bmatrix} 2 \\ 1 \\ 2 \end{bmatrix}, \qquad q_3 = \frac{1}{3} \begin{bmatrix} 2 \\ -2 \\ -1 \end{bmatrix}. \] Find \(q_3\) by finding the left nullspace of \(A\) and then normalizing, or by taking \(q_1 \times q_2\), or by guessing a vector independent of \(q_1\) and \(q_2\) and using Gram-Schmidt. 1b. \(q_3\) is in the left nullspace of \(A\), since it is orthogonal to both columns of \(A\). 1c. \[ P = q_3(q_3^T q_3)^{-1} q_3^T = \frac{1}{9} \begin{bmatrix} 4 & -4 & -2 \\ -4 & 4 & 2 \\ -2 & 2 & 1 \end{bmatrix}. \] 1d. \[ \hat{x} = (A^T A)^{-1} A^T \vec{b} = \begin{bmatrix} 1 \\ 2 \end{bmatrix}. \]

Keywords: orthonormal basis, left nullspace, projection matrix, least squares