Skip to content

6.2-final-1998

Final Exam, 1998 · Section 6.2: Diagonalizing a Matrix Source: finalexam_1.pdf, pp. 9, 12

Problem

8. (a) (5 points) Find a complete set of eigenvalues and eigenvectors for the matrix

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

(b) (6 points, 1 each) Circle all the properties of this matrix \(A\):

  • \(A\) is a projection matrix
  • \(A\) is a positive definite matrix
  • \(A\) is a Markov matrix
  • \(A\) has determinant larger than trace
  • \(A\) has three orthonormal eigenvectors
  • \(A\) can be factored into \(A = LU\)

(c) (4 points) Write the vector \(u_0 = \begin{bmatrix} 2 \\ 0 \\ 1 \end{bmatrix}\) as a combination of eigenvectors of \(A\), and compute the vector \(u_{100} = A^{100}u_0\).

Solution (a) The eigenvalues are \(\lambda = 1, 1, 4\). Eigenvectors can be \[ \begin{bmatrix} 1 \\ -1 \\ 0 \end{bmatrix}, \quad \begin{bmatrix} 1 \\ 0 \\ -1 \end{bmatrix}, \quad \begin{bmatrix} 1 \\ 1 \\ 1 \end{bmatrix} \] (for \(\lambda = 1, 1, 4\), respectively; they could also be chosen orthonormal because \(A = A^T\)). (b) The properties that hold: \(A\) is a positive definite matrix (eigenvalues \(1, 1, 4\) are all positive); \(A\) has three orthonormal eigenvectors (it is symmetric); \(A\) can be factored into \(A = LU\) (pivots \(2, \frac{3}{2}, \frac{4}{3}\) are nonzero, so no row exchanges are needed). The properties that do not hold: \(A\) is not a projection matrix (its eigenvalues must be 0 or 1, but 4 is an eigenvalue, i.e. \(A^2 \neq A\)); \(A\) is not a Markov matrix (each row sums to 4, not 1); \(\det A = 1 \cdot 1 \cdot 4 = 4\) is not larger than the trace \(2 + 2 + 2 = 6\). (c) \[ \begin{bmatrix} 2 \\ 0 \\ 1 \end{bmatrix} = \begin{bmatrix} 1 \\ -1 \\ 0 \end{bmatrix} + \begin{bmatrix} 1 \\ 1 \\ 1 \end{bmatrix}, \] \[ A^{100}\begin{bmatrix} 2 \\ 0 \\ 1 \end{bmatrix} = 1^{100} \begin{bmatrix} 1 \\ -1 \\ 0 \end{bmatrix} + 4^{100} \begin{bmatrix} 1 \\ 1 \\ 1 \end{bmatrix} = \begin{bmatrix} 4^{100} + 1 \\ 4^{100} - 1 \\ 4^{100} \end{bmatrix}. \]

Keywords: eigenvalues, eigenvectors, matrix powers, positive definite