Skip to content

3.5-exam2-1998-b

Quiz 2, 1998 · Section 3.5: Dimensions of the Four Subspaces Source: exam2_1.pdf, pp. 17, 18

Problem

4. (30 pts) Let \(A\) be an \(n \times n\) matrix.

(a.) If the row space of \(A\) is \(R^n\) then the column space of \(A\) is ________ ?

(b.) If the nullspace of \(A\) is \(R^n\) then the column space of \(A\) is ________ ?

(c.) If the left nullspace of \(A\) is \(R^n\) then the column space of \(A\) is ________ ?

The next three questions consider a square matrix \(A\) whose column space is orthogonal to the row space.

(d.) Give an example of a square matrix \(A\) such that the column space is orthogonal to the row space.

(e.) If the column space of an \(n \times n\) matrix \(A\) is orthogonal to the row space there is an inequality relating the rank \(r\) to \(n\). What is the strongest possible inequality? (Hint: \(r \leq n\) is a true inequality, but is not the strongest and hence will be considered an incorrect answer. Only the right answer will be given credit.)

(f.) If the column space is orthogonal to the row space, then \(\det(A) =\) ________ ?

Solution 4a. \(R(A) = R^n \Rightarrow \text{rank}(A) = n \Rightarrow \dim C(A) = n \Rightarrow C(A) = R^n\). 4b. \(N(A) = R^n \Rightarrow \text{rank}(A) = 0 \Rightarrow C(A) = \{0\}\), the zero vector space. 4c. \(N(A^T) = R^n \Rightarrow \text{rank}(A) = 0 \Rightarrow C(A) = \{0\}\). 4d. Example 1: \(A = [0]\). Example 2: \[ A = \begin{bmatrix} 0 & 0_m \\ I_k & 0 \end{bmatrix} \] where \(I_k\) is the \(k \times k\) identity matrix and \(0_m\) is the \(m \times m\) zero matrix. 4e. \[ \left. \begin{array}{l} C(A) \perp R(A) \\ N(A) = (R(A))^{\perp} \end{array} \right\} \Rightarrow C(A) \subseteq N(A) \] Hence \[ r + r = \dim C(A) + \dim R(A) \leq \dim N(A) + \dim R(A) = n, \] i.e. \(2r \leq n\). Example 2 above with \(k = \lceil \frac{n}{2} \rceil\), \(m = n - k\) shows that \(2r \leq n\) is the strongest possible inequality. 4f. From (e), \(r \leq \frac{n}{2} < n\), so \(A\) is singular and \(\det(A) = 0\).

Keywords: four subspaces, rank, orthogonal column space, determinant