5.2-exam2-1999¶
Quiz 2, 1999 · Section 5.2: Computing and Using Determinants Source: exam2_1.pdf, pp. 25, 29
Problem¶
1 (30 pts.) (a) Compute the determinant of
\[
A = \begin{bmatrix} 1 & -1 & 1 & 0 \\ 1 & 1 & 5 & 0 \\ 1 & 3 & 9 & 0 \\ 0 & 0 & 0 & 2 \end{bmatrix}.
\]
(b) Find an orthogonal basis (orthonormal is even better) for the column space of \(A\). Start from a basis and use Gram-Schmidt (and common sense).
(c) If you change the 1 in the upper left corner of \(A\) to 2, what is the change in the determinant (I would use cofactors).
Solution
1a. \(|A| = 2\) times the 3 by 3 determinant \(= 2\) times \(0 = 0\). 1b. \(A\) has rank 3, so we want three orthogonal basis vectors \(A, B, C\): \[ A = \text{first column } (1, 1, 1, 0) \] \[ B = (\text{second column}) - (\text{projection onto first column}) = (-1, 1, 3, 0) - (1, 1, 1, 0) \cdot \frac{3}{3} = (-2, 0, 2, 0) \quad \text{(check: orthogonal to first column)} \] \[ C = \text{last column } (0, 0, 0, 2) \] To orthonormalize, divide by the lengths: \[ q_1 = \frac{A}{\|A\|} = \frac{(1, 1, 1, 0)}{\sqrt{3}}, \qquad q_2 = \frac{B}{2\sqrt{2}} = \frac{(-1, 0, 1, 0)}{\sqrt{2}}, \qquad q_3 = \frac{C}{2} = (0, 0, 0, 1). \] 1c. Adding 1 to the \(a_{11}\) entry will add its cofactor to the determinant: \[ C_{11} = \begin{vmatrix} 1 & 5 & 0 \\ 3 & 9 & 0 \\ 0 & 0 & 2 \end{vmatrix} = -12, \] so the determinant changes by \(-12\).Keywords: determinant, Gram-Schmidt, orthogonal basis, cofactors