Skip to content

8.1-final-2011-strang

Final Exam, 2011 (strang) · Section 8.1: The Idea of a Linear Transformation Source: finalexam_2.pdf, p. 56

Problem

9. (a) The index of a matrix \(A\) is the dimension of its nullspace minus the dimension of the nullspace of \(A^T\). If \(A\) is a 9 by 7 matrix of rank \(r\), what is its index?

(b) Suppose \(M\) is the vector space consisting of all 2 by 2 matrices. (So those matrices are the "vectors" in \(M\).) Write down a basis for this vector space \(M\): linearly independent and spanning the space \(M\).

(c) \(S = \begin{bmatrix} 1 & 0 \\ 1 & 1 \end{bmatrix}\) is a specific matrix in \(M\). For every 2 by 2 matrix \(A\), the transformation \(T\) produces \(T(A) = S^{-1}AS\). Is this a linear transformation? What tests do you have to check?

Solution (a) The nullspace of \(A\) has dimension \(n - r = 7 - r\) and the nullspace of \(A^T\) has dimension \(m - r = 9 - r\). The index is \[ (7 - r) - (9 - r) = -2. \] (b) A basis for the 2 by 2 matrices is \[ \begin{bmatrix} 1 & 0 \\ 0 & 0 \end{bmatrix}, \quad \begin{bmatrix} 0 & 1 \\ 0 & 0 \end{bmatrix}, \quad \begin{bmatrix} 0 & 0 \\ 1 & 0 \end{bmatrix}, \quad \begin{bmatrix} 0 & 0 \\ 0 & 1 \end{bmatrix}. \] They are independent (no nonzero combination gives the zero matrix) and every 2 by 2 matrix is a linear combination of them. (c) Yes, \(T\) is a linear transformation. The two tests to check are \[ T(A_1 + A_2) = T(A_1) + T(A_2), \qquad T(cA) = c\,T(A), \] for all 2 by 2 matrices \(A, A_1, A_2\) and all scalars \(c\). Both hold here because matrix multiplication distributes over addition and over scalar multiplication: \[ S^{-1}(A_1 + A_2)S = S^{-1}A_1S + S^{-1}A_2S, \qquad S^{-1}(cA)S = c\,S^{-1}AS. \]

Keywords: index, vector space of matrices, linear transformation, basis