6.3-exam3-2000-spring¶
Exam 3, 2000 (spring) · Section 6.3: Symmetric Positive Definite Matrices Source: exam3_1.pdf, pp. 49, 52
Problem¶
2 (25 pts) True (give a reason) or False (give a counterexample):
(a) If \(A\) is a symmetric matrix, any two eigenvectors of \(A\) are perpendicular.
(b) If \(A\) is \(n \times n\) and has \(n\) orthonormal eigenvectors, then \(A\) is symmetric.
(c) Any eigenvector matrix \(S\) of a symmetric matrix is symmetric.
Note: You must show your work to receive credit for this problem.
Solution
(a) False. The \(2 \times 2\) identity matrix is symmetric, but it has plenty of non-perpendicular eigenvectors, e.g., \((1, 0)\) and \((1, 1)\). (b) True, because \(A\) is diagonalizable using an orthogonal matrix: \(A = Q\Lambda Q^T\). Such a matrix is symmetric: \((Q\Lambda Q^T)^T = Q\Lambda Q^T\). (c) False. Same example as in part (a): the identity matrix is symmetric but its eigenvector matrix \(S = \begin{bmatrix} 1 & 1 \\ 0 & 1 \end{bmatrix}\) is not symmetric.Keywords: symmetric matrix, eigenvectors, orthonormal, spectral theorem