Skip to content

6.5-exam3-2012

Quiz 3, 2012 · Section 6.5: Solving Linear Differential Equations Source: exam3_2.pdf, pp. 37-38

Problem

2. (30 pts.) Let \(A = \begin{bmatrix} 0 & 1 & 1 \\ 0 & 0 & 1 \\ 0 & 0 & 0 \end{bmatrix}\).

a) (6 pts.) What are the eigenvalues of \(A\)? (Explain briefly.)

b) (6 pts.) What is the rank of \(A\)?

c) (6 pts.) What are the singular values of \(A\)?

d) (6 pts.) What is the Jordan form of \(A\)? (Explain briefly.)

e) (6 pts.) Compute in simplest form \(e^{tA}\).

Solution a) The matrix \(A\) is upper triangular. For such a matrix, the determinant is the product of the diagonal entries. Using this observation, if we try to compute \(|A - xI|\), we find \(-x^3\). This implies that the only eigenvalue is 0 with multiplicity 3. b) It is clear that the last two columns of \(A\) are pivot columns. Therefore, the rank is 2. c) The singular values of \(A\) are the square roots of the eigenvalues of \(A^TA\): \[ A^TA = \begin{bmatrix} 0 & 0 & 0 \\ 0 & 1 & 1 \\ 0 & 1 & 2 \end{bmatrix}. \] We have \[ |xI - A^TA| = x((x - 1)(x - 2) - 1) = x(x^2 - 3x + 1), \] whose roots are \(0, \frac{3 + \sqrt{5}}{2}, \frac{3 - \sqrt{5}}{2}\). Therefore, the singular values of \(A\) are \(0, \sqrt{\frac{3 + \sqrt{5}}{2}}\), and \(\sqrt{\frac{3 - \sqrt{5}}{2}}\). d) In general, the Jordan form has zeroes everywhere except on the diagonal, where you put the eigenvalues, and on the second diagonal, where you have 1s and 0s. Note that the matrix \(A\) as it is is not in Jordan normal form because you have a 1 in the upper right corner. There are 3 possibilities for what the Jordan form can be: one with two ones over the diagonal and two with one one and one zero. To determine which is the actual Jordan form, you can look at the rank. We know that \(A\) has rank 2 and the Jordan form is similar to \(A\), so it must have rank 2 as well. Therefore, the only possibility is \[ \begin{bmatrix} 0 & 1 & 0 \\ 0 & 0 & 1 \\ 0 & 0 & 0 \end{bmatrix}. \] e) We can use the series expression for \(e^{tA}\). In general this is an infinite sum which is unpleasant, but in this particular case the powers of \(A\) quickly become zero. Indeed, we have \[ A^2 = \begin{bmatrix} 0 & 0 & 1 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \end{bmatrix}, \qquad A^3 = 0. \] Therefore, we have \[ e^{tA} = I + tA + \frac{t^2}{2}A^2 = \begin{bmatrix} 1 & t & t + t^2/2 \\ 0 & 1 & t \\ 0 & 0 & 1 \end{bmatrix}. \]

Keywords: nilpotent matrix, eigenvalues, singular values, Jordan form, matrix exponential