Skip to content

5.2-exam2-2012

Quiz 2, 2012 · Section 5.2: Computing and Using Determinants Source: exam2_2.pdf, pp. 16-20

Problem

2 (22 pts.)

Consider a 4 × 4 matrix

\[ A = \begin{bmatrix} 0 & x & y & z \\ x & 1 & 0 & 0 \\ y & 0 & 1 & 0 \\ z & 0 & 0 & 1 \end{bmatrix}. \]

a) (17 pts.) Compute \(|A|\), the determinant of \(A\), in simplest form.

b) (5 pts.) For what values of \(x, y, z\) is \(A\) singular?

Solution a) The answer is \(\det A = -x^2 - y^2 - z^2\). Before computing, note that this expression is symmetric in the three variables \(x, y, z\). We might have predicted this: if we swap rows 2 and 3 of \(A\), and then swap columns 2 and 3 of the result, we end up with \[ A' = \begin{bmatrix} 0 & y & x & z \\ y & 1 & 0 & 0 \\ x & 0 & 1 & 0 \\ z & 0 & 0 & 1 \end{bmatrix}, \] which is the same as \(A\), but with the roles of \(x\) and \(y\) swapped. In performing one row swap and one column swap, we have multiplied the determinant by \((-1)^2 = 1\), so \(A'\) has the same determinant as \(A\). From this we conclude that \(\det A\) must be an expression symmetric in \(x\) and \(y\); similar considerations show that it is symmetric in all three variables. Here are some of the most common ways from the students' tests to compute \(\det A\): - By cofactor expansion in the first row, using the big formula or any other method for each 3 × 3 minor: \[ \det A = 0 \begin{vmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{vmatrix} - x \begin{vmatrix} x & 0 & 0 \\ y & 1 & 0 \\ z & 0 & 1 \end{vmatrix} + y \begin{vmatrix} x & 1 & 0 \\ y & 0 & 0 \\ z & 0 & 1 \end{vmatrix} - z \begin{vmatrix} x & 1 & 0 \\ y & 0 & 1 \\ z & 0 & 0 \end{vmatrix} = 0 - x(x) + y(-y) - z(z) = -x^2 - y^2 - z^2. \] Note the alternating signs in the cofactors; in general \(C_{ij} = (-1)^{i+j}\det M_{ij}\). - By cofactor expansion in the second row: \[ \det A = -x \begin{vmatrix} x & y & z \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{vmatrix} + 1 \begin{vmatrix} 0 & y & z \\ y & 1 & 0 \\ z & 0 & 1 \end{vmatrix} - 0 \begin{vmatrix} 0 & x & z \\ y & 0 & 0 \\ z & 0 & 1 \end{vmatrix} + 0 \begin{vmatrix} 0 & x & y \\ y & 0 & 1 \\ z & 0 & 0 \end{vmatrix} = -x(1) + 1(-y^2 - z^2) = -x^2 - y^2 - z^2. \] The cofactor \(C_{22}\), for example, can be calculated with the big formula for 3 × 3 matrices: \[ \begin{vmatrix} 0 & y & z \\ y & 1 & 0 \\ z & 0 & 1 \end{vmatrix} = 0 \cdot 1 \cdot 1 + y \cdot 0 \cdot z + z \cdot y \cdot 0 - 0 \cdot 0 \cdot 0 - y \cdot y \cdot 1 - z \cdot 1 \cdot z = -y^2 - z^2. \] - By the big formula for 4 × 4 matrices. The big formula has 24 terms (one for each 4 × 4 permutation matrix), but only three of them are nonzero: \[ \det A = x^2 \det \begin{bmatrix} 0 & 1 & 0 & 0 \\ 1 & 0 & 0 & 0 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 1 \end{bmatrix} + y^2 \det \begin{bmatrix} 0 & 0 & 1 & 0 \\ 0 & 1 & 0 & 0 \\ 1 & 0 & 0 & 0 \\ 0 & 0 & 0 & 1 \end{bmatrix} + z^2 \det \begin{bmatrix} 0 & 0 & 0 & 1 \\ 0 & 1 & 0 & 0 \\ 0 & 0 & 1 & 0 \\ 1 & 0 & 0 & 0 \end{bmatrix}. \] These three permutation matrices all have determinant \(-1\), because they are one row exchange away from the identity matrix, so \(\det A = -x^2 - y^2 - z^2\). - By row operations to reach an upper triangular matrix. First exchange row 1 with row 4 to put a pivot in the top-left corner (row exchanges negate the determinant): \[ |A| = -\begin{vmatrix} z & 0 & 0 & 1 \\ x & 1 & 0 & 0 \\ y & 0 & 1 & 0 \\ 0 & x & y & z \end{vmatrix}. \] Now subtract \(x/z\) times row 1 from row 2, and \(y/z\) times row 1 from row 3 (such operations do not affect the determinant): \[ |A| = -\begin{vmatrix} z & 0 & 0 & 1 \\ 0 & 1 & 0 & -x/z \\ 0 & 0 & 1 & -y/z \\ 0 & x & y & z \end{vmatrix}. \] Finally subtract \(x\) times row 2 and \(y\) times row 3 from row 4: \[ |A| = -\begin{vmatrix} z & 0 & 0 & 1 \\ 0 & 1 & 0 & -x/z \\ 0 & 0 & 1 & -y/z \\ 0 & 0 & 0 & z + x^2/z + y^2/z \end{vmatrix}. \] Multiplying the diagonal entries gives \[ |A| = -z \cdot 1 \cdot 1 \cdot \left( z + \frac{x^2}{z} + \frac{y^2}{z} \right) = -x^2 - y^2 - z^2. \] - By row operations to reach a lower triangular matrix. From row 1 of \(A\), subtract \(x\) times row 2, \(y\) times row 3, and \(z\) times row 4. These operations do not change the determinant, so \[ |A| = \begin{vmatrix} -x^2 - y^2 - z^2 & 0 & 0 & 0 \\ x & 1 & 0 & 0 \\ y & 0 & 1 & 0 \\ z & 0 & 0 & 1 \end{vmatrix} = -x^2 - y^2 - z^2. \] In other words, we may factorize \(A\) as \[ A = \begin{bmatrix} 1 & x & y & z \\ 0 & 1 & 0 & 0 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 1 \end{bmatrix} \begin{bmatrix} -x^2 - y^2 - z^2 & 0 & 0 & 0 \\ x & 1 & 0 & 0 \\ y & 0 & 1 & 0 \\ z & 0 & 0 & 1 \end{bmatrix}, \] so the product rule says \(|A| = -x^2 - y^2 - z^2\). b) A square matrix is singular if and only if its determinant equals zero. So we are asked to find all triples \((x, y, z)\) such that \[ \det A = -x^2 - y^2 - z^2 = 0, \quad \text{or in other words} \quad x^2 + y^2 + z^2 = 0. \] For real numbers \(x, y, z\), the left-hand side is just the square of the distance from \((x, y, z)\) to the origin in \(R^3\). Since only the origin is at distance 0 from the origin, the matrix \(A\) is singular if and only if \(x = y = z = 0\).

Keywords: determinant, cofactor expansion, big formula, symmetric matrix