Skip to content

6.2-exam3-2009-spring-b

Quiz 3, 2009 (spring) · Section 6.2: Diagonalizing a Matrix Source: exam3_1.pdf, p. 161

Problem

Problem 3:

The \(3 \times 3\) matrix \(A\) has three independent eigenvectors \(v_1\), \(v_2\), and \(v_3\) with corresponding eigenvalues \(\lambda_1\), \(\lambda_2\), and \(\lambda_3\) (that is, \(Av_i = \lambda_iv_i\) for \(i = 1, 2, 3\)).

If

\[ b = c_1v_1 + c_2v_2 + c_3v_3 \]

for some coefficients \(c_1\), \(c_2\), and \(c_3\), then write (in terms of \(\lambda_i\), \(c_i\), and \(v_i\)) a formula for the solution \(x\) of

\[ A^2x + 2Ax - 3Ix = b \]

(you can assume that a solution exists for any \(b\)).

Solution Using the eigenvectors \(v_1\), \(v_2\), \(v_3\), we have \[ x = (A^2 + 2A - 3I)^{-1}b \] \[ = (A^2 + 2A - 3I)^{-1}(c_1v_1 + c_2v_2 + c_3v_3) \] \[ = \frac{c_1}{\lambda_1^2 + 2\lambda_1 - 3}v_1 + \frac{c_2}{\lambda_2^2 + 2\lambda_2 - 3}v_2 + \frac{c_3}{\lambda_3^2 + 2\lambda_3 - 3}v_3. \]

Keywords: eigenvalues, eigenvectors, matrix polynomial, inverse matrix