7.1-final-2007¶
Final Exam, 2007 · Section 7.1: Singular Values and Singular Vectors Source: finalexam_1.pdf, p. 174
Problem¶
Problem 5 (10 points) Let \(A = \begin{bmatrix} 1 & 1 & 1 & 1 \\ 2 & 2 & 2 & 2 \\ 3 & 3 & 3 & 3 \end{bmatrix}\).
(a) What is the rank of \(A\)?
(b) Calculate the matrix \(A^TA\). Find all its eigenvalues (with multiplicities).
(c) Calculate the matrix \(AA^T\). Find all its eigenvalues (with multiplicities).
(d) Find the matrix \(\Sigma\) in the singular value decomposition \(A = U\Sigma V^T\).
Solution
(a) 1 (b) \[ A^TA = \begin{bmatrix} 14 & 14 & 14 & 14 \\ 14 & 14 & 14 & 14 \\ 14 & 14 & 14 & 14 \\ 14 & 14 & 14 & 14 \end{bmatrix}. \] It was rank 1 so the eigenvalues are 56, 0, 0, 0. (c) \[ AA^T = \begin{bmatrix} 4 & 8 & 12 \\ 8 & 16 & 24 \\ 12 & 24 & 36 \end{bmatrix}. \] It was rank 1 so the eigenvalues are 56, 0, 0. (d) \[ \Sigma = \begin{bmatrix} \sqrt{56} & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \end{bmatrix}. \]Keywords: singular value decomposition, rank one matrix, eigenvalues, normal equations