Skip to content

6.2-exam3-2000-spring

Exam 3, 2000 (spring) · Section 6.2: Diagonalizing a Matrix Source: exam3_1.pdf, pp. 51, 52

Problem

4 (25 pts) Suppose \(A\) is a 3 \(\times\) 3 matrix with eigenvalues \(\lambda = 1\) and \(\lambda = 2\). Suppose also that \(A - I\) has rank one.

(a) Which eigenvalue of \(A\) is repeated? Explain why.

(b) Write down a specific matrix which is similar to \(A\) and symmetric.

Explain why they are similar.

(c) Write down a specific matrix which is similar to \(A\) and not symmetric.

Explain why they are similar.

(d) Write down a specific matrix which has the same eigenvalues as \(A\) but is not similar to \(A\). Explain why they are not similar.

Solution (a) \(\lambda = 1\) is repeated. The number of independent \(\lambda = 1\) eigenvectors is given by the dimension of \(N(A - I)\), which is \(3 - rank(A - I) = 3 - 1 = 2\). So \(A\) has two independent \(\lambda = 1\) eigenvectors, and therefore \(\lambda = 1\) must be the repeated eigenvalue. (b) \(A\) has three independent eigenvectors, so it is diagonalizable, i.e., similar to \[ \Lambda = \begin{bmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 2 \end{bmatrix}, \] which is symmetric. The matrices are similar because both diagonalize to the same \(\Lambda\). (c) The matrix \[ B = \begin{bmatrix} 1 & 0 & 0 \\ 0 & 1 & 1 \\ 0 & 0 & 2 \end{bmatrix} \] has the same eigenvalues and the same number of independent eigenvectors as \(A\), so it is similar to \(A\). (d) The matrix \[ C = \begin{bmatrix} 1 & 1 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 2 \end{bmatrix} \] has the same eigenvalues as \(A\) but is missing an eigenvector: the rank of \(C - I\) is two, so \(C\) has only one independent \(\lambda = 1\) eigenvector. Hence \(C\) is not similar to \(A\).

Keywords: eigenvalues, similar matrices, diagonalizable, rank