Skip to content

6.2-exam3-2001-b

Quiz 3, 2001 · Section 6.2: Diagonalizing a Matrix Source: exam3_1.pdf, p. 56

Problem

3 (32 pts.) Let

\[ A = \begin{bmatrix} 5 & 4 & 3 \\ -1 & 0 & -3 \\ 1 & -2 & 1 \end{bmatrix} \]

and

\[ X = \begin{bmatrix} 1 & 1 & 1 \\ -1 & -1 & 0 \\ -1 & 1 & 0 \end{bmatrix} \quad \text{so that} \quad X^{-1} = \begin{bmatrix} 0 & -\frac{1}{2} & -\frac{1}{2} \\ 0 & -\frac{1}{2} & \frac{1}{2} \\ 1 & 1 & 0 \end{bmatrix}. \]

(a) Compute \(M = X^{-1}AX\). (4p)

What are the eigenvalues of \(A\)? (4p)

How many linearly independent eigenvectors does \(A\) have? (4p)

Is \(A\) diagonalizable? (4p)

(b) Let

\[ B = \begin{bmatrix} -3 & 0 & 0 \\ 0 & 0 & 1 \\ 0 & 0 & 0 \end{bmatrix} \quad \text{and} \quad x = \begin{bmatrix} 1 \\ 0 \\ 0 \end{bmatrix}. \]

Compute \(e^{Bt}\) explicitly. (12p)

Compute \(\lim_{t \to \infty} e^{Bt}x\). (4p)

Solution (a) Multiplying out, \[ AX = \begin{bmatrix} -2 & 4 & 5 \\ 2 & -4 & -1 \\ 2 & 4 & 1 \end{bmatrix}, \] so \[ M = X^{-1}AX = \begin{bmatrix} 0 & -\frac{1}{2} & -\frac{1}{2} \\ 0 & -\frac{1}{2} & \frac{1}{2} \\ 1 & 1 & 0 \end{bmatrix} \begin{bmatrix} -2 & 4 & 5 \\ 2 & -4 & -1 \\ 2 & 4 & 1 \end{bmatrix} = \begin{bmatrix} -2 & 0 & 0 \\ 0 & 4 & 1 \\ 0 & 0 & 4 \end{bmatrix}. \] The eigenvalues of \(A\) are those of \(M\), an upper triangular matrix: \(-2, 4, 4\). The eigenvalue \(4\) of \(M\) has only one independent eigenvector (since \(M - 4I = \begin{bmatrix} -6 & 0 & 0 \\ 0 & 0 & 1 \\ 0 & 0 & 0 \end{bmatrix}\) has rank 2), so counting the eigenvector for \(-2\), \(A\) has 2 linearly independent eigenvectors in total. Since there are only 2 independent eigenvectors in 3 dimensions, \(A\) is not diagonalizable. (b) The matrix \(B\) is block diagonal: \(-3\) in the \(1 \times 1\) block and the nilpotent \(N = \begin{bmatrix} 0 & 1 \\ 0 & 0 \end{bmatrix}\) in the \(2 \times 2\) block. Since \(N^2 = 0\), \[ e^{Bt} = \begin{bmatrix} e^{-3t} & 0 & 0 \\ 0 & 1 & t \\ 0 & 0 & 1 \end{bmatrix}. \] With \(x = (1, 0, 0)\), \[ e^{Bt}x = \begin{bmatrix} e^{-3t} \\ 0 \\ 0 \end{bmatrix} \to \begin{bmatrix} 0 \\ 0 \\ 0 \end{bmatrix} \quad \text{as } t \to \infty. \]

Keywords: similar matrices, eigenvalues, diagonalizable, matrix exponential