Orthogonal Matrices - Linear Algebra

  1. Why is [[2,0],[0,3]][[2, 0], [0, 3]] not an orthogonal matrix, although its columns are orthogonal?
    Its columns are not unit vectors: QTQ=[[4,0],[0,9]]≠IQ^TQ = [[4, 0], [0, 9]] \ne I.
  2. What can be said about the rows of an orthogonal matrix?
    They are also orthonormal, because QTQ^T is orthogonal too (QQT=IQQ^T = I).
  3. What are the possible values of the determinant of an orthogonal matrix?
    Only 11 and −1-1, since (det⁡Q)2=det⁡(QTQ)=1(\det Q)^2 = \det(Q^TQ) = 1.
  4. QQ is a 3×33 \times 3 orthogonal matrix and ∥x∥=7\|\mathbf{x}\| = 7. What is ∥Qx∥\|Q\mathbf{x}\|?
    77. Orthogonal matrices preserve length.
  5. For orthogonal Q1Q_1 and Q2Q_2 of the same size, what is (Q1Q2)T(Q1Q2)(Q_1Q_2)^T(Q_1Q_2), and what does it show?
    II, since Q2TQ1TQ1Q2=Q2TQ2=IQ_2^TQ_1^TQ_1Q_2 = Q_2^TQ_2 = I. So a product of orthogonal matrices is orthogonal.
  6. What is the inverse of an orthogonal matrix QQ?
    Q−1=QTQ^{-1} = Q^T.
  7. What does multiplication by an orthogonal matrix QQ preserve?
    Lengths and dot products: ∥Qx∥=∥x∥\|Q\mathbf{x}\| = \|\mathbf{x}\| and (Qx)⋅(Qy)=x⋅y(Q\mathbf{x}) \cdot (Q\mathbf{y}) = \mathbf{x} \cdot \mathbf{y}, so angles too.
  8. What is the inverse of the rotation matrix [[cos⁡θ,−sin⁡θ],[sin⁡θ,cos⁡θ]][[\cos\theta, -\sin\theta], [\sin\theta, \cos\theta]]?
    Its transpose [[cos⁡θ,sin⁡θ],[−sin⁡θ,cos⁡θ]][[\cos\theta, \sin\theta], [-\sin\theta, \cos\theta]], the rotation by −θ-\theta.
  9. A 2×22 \times 2 orthogonal matrix has determinant −1-1. What kind of transformation is it?
    A reflection across a line through the origin.
  10. What is an orthogonal matrix?
    A square matrix QQ with QTQ=IQ^TQ = I, that is, with orthonormal columns.
  11. What value of aa makes [[35,a],[45,35]][[\frac{3}{5}, a], [\frac{4}{5}, \frac{3}{5}]] an orthogonal matrix?
    a=−45a = -\frac{4}{5}. Column 2 must be orthogonal to (35,45)(\frac{3}{5}, \frac{4}{5}): 3a5+1225=0\frac{3a}{5} + \frac{12}{25} = 0; its length is then 11.
  12. Which real numbers can be eigenvalues of an orthogonal matrix?
    Only 11 and −1-1, because QQ preserves length, so every eigenvalue has ∣λ∣=1|\lambda| = 1.
  13. What is the inverse of the permutation matrix [[0,1],[1,0]][[0, 1], [1, 0]]?
    The matrix itself. It is orthogonal and symmetric, so its inverse equals its transpose, which is itself.

Study these flashcards