1.

4 Inverses; Algebraic Properties of Matrices 51

EXAMPLE 13 | Inverse of a Transpose

Consider a general 2 × 2 invertible matrix and its transpose:

a b a c
𝐴=[ ] and 𝐴𝑇 = [ ]
c d b d
Since 𝐴 is invertible, its determinant ad − bc is nonzero. But the determinant of 𝐴𝑇 is also
ad − bc (verify), so 𝐴𝑇 is also invertible. It follows from Theorem 1.4.5 that
d c
⎡ ad − bc −
ad − bc ⎤
(𝐴𝑇 )−1 =⎢ ⎥
⎢ b a ⎥
⎣− ad − bc ad − bc ⎦
which is the same matrix that results if 𝐴−1 is transposed (verify). Thus,
(𝐴𝑇 )−1 = (𝐴−1 )𝑇
as guaranteed by Theorem 1.4.9.

Exercise Set 1.4

In Exercises 1–2, verify that the following matrices and scalars satisfy 9. Find the inverse of
the stated properties of Theorem 1.4.1. 1 x 1 x
2 (e + e−x ) 2 (e − e−x )
[ ]
3 −1 0 2 1 x
− e−x ) 1 x
+ e−x )
𝐴=[ ], 𝐵=[ ], 2 (e 2 (e
2 4 1 −4
10. Find the inverse of
4 1
𝐶=[ ], a = 4, b = −7 cos 𝜃 sin 𝜃
−3 −2 [ ]
− sin 𝜃 cos 𝜃
1. a. The associative law for matrix addition. In Exercises 11–14, verify that the equations are valid for the matri-
b. The associative law for matrix multiplication. ces in Exercises 5–8.
c. The left distributive law. 11. (𝐴𝑇 )−1 = (𝐴−1 )𝑇 12. (𝐴−1 )−1 = 𝐴

d. (a + b)𝐶 = a𝐶 + b𝐶 13. (𝐴𝐵𝐶)−1 = 𝐶 −1 𝐵 −1 𝐴−1 14. (𝐴𝐵𝐶)𝑇 = 𝐶 𝑇𝐵 𝑇𝐴𝑇

2. a. a(𝐵𝐶) = (a𝐵)𝐶 = 𝐵(a𝐶) In Exercises 15–18, use the given information to find 𝐴.
b. 𝐴(𝐵 − 𝐶) = 𝐴𝐵 − 𝐴𝐶 c. (𝐵 + 𝐶)𝐴 = 𝐵𝐴 + 𝐶𝐴 −3 7 −3 −1
15. (7𝐴)−1 = [ ] 16. (5𝐴𝑇 )−1 = [ ]
1 −2 5 2
d. a(b𝐶) = (ab)𝐶
−1 2 2 −1
In Exercises 3–4, verify that the matrices and scalars in Exercise 1 17. (𝐼 + 2𝐴)−1 = [ ] 18. 𝐴−1 = [ ]
4 5 3 5
satisfy the stated properties.
In Exercises 19–20, compute the following using the given matrix 𝐴.
3. a. (𝐴𝑇 )𝑇 = 𝐴 b. (𝐴𝐵)𝑇 = 𝐵 𝑇𝐴𝑇
a. 𝐴3 b. 𝐴−3 c. 𝐴2 − 2𝐴 + 𝐼
4. a. (𝐴 + 𝐵)𝑇 = 𝐴𝑇 + 𝐵 𝑇 b. (a𝐶)𝑇 = a𝐶 𝑇
3 1 2 0
19. 𝐴 = [ ] 20. 𝐴 = [ ]
In Exercises 5–8, use Theorem 1.4.5 to compute the inverse of the 2 1 4 1
matrix.
In Exercises 21–22, compute p(𝐴) for the given matrix 𝐴 and the
2 −3 3 1 following polynomials.
5. 𝐴 = [ ] 6. 𝐵 = [ ]
4 4 5 2
a. p(x) = x − 2
2 0 6 4 b. p(x) = 2x 2 − x + 1
7. 𝐶 = [ ] 8. 𝐷 = [ ]
0 3 −2 −1 c. p(x) = x 3 − 2x + 1
52 CH APT ER 1 Systems of Linear Equations and Matrices

3 1 2 0 36. Can a matrix with two identical rows or two identical columns
21. 𝐴 = [ ] 22. 𝐴 = [ ] have an inverse? Explain.
2 1 4 1

In Exercises 23–24, let In Exercises 37–38, determine whether 𝐴 is invertible, and if so, find
the inverse. [Hint: Solve 𝐴𝑋 = 𝐼 for 𝑋 by equating corresponding
a b 0 1 0 0 entries on the two sides.]
𝐴=[ ], 𝐵=[ ], 𝐶=[ ]
c d 0 0 1 0 1 0 1 1 1 1
37. 𝐴 = [1 1 0] 38. 𝐴 = [1 0 0]
23. Find all values of a, b, c, and d (if any) for which the matrices 0 1 1 0 1 1
𝐴 and 𝐵 commute.
In Exercises 39–40, simplify the expression assuming that 𝐴, 𝐵, 𝐶,
24. Find all values of a, b, c, and d (if any) for which the matrices and 𝐷 are invertible.
𝐴 and 𝐶 commute. 39. (𝐴𝐵)−1 (𝐴𝐶 −1 )(𝐷 −1 𝐶 −1 )−1 𝐷 −1
In Exercises 25–28, use the method of Example 8 to find the unique
40. (𝐴𝐶 −1 )−1 (𝐴𝐶 −1 )(𝐴𝐶 −1 )−1 𝐴𝐷 −1
solution of the given linear system.
41. Show that if 𝑅 is a 1 × n matrix and 𝐶 is an n × 1 matrix, then
25. 3x 1 − 2x 2 = −1 26. −x 1 + 5x 2 = 4 𝑅𝐶 = tr(𝐶𝑅).
4x 1 + 5x 2 = 3 −x 1 − 3x 2 = 1
42. If 𝐴 is a square matrix and n is a positive integer, is it true that
27. 6x 1 + x 2 = 0 28. 2x 1 − 2x 2 = 4 (𝐴n )𝑇 = (𝐴𝑇 )n ? Justify your answer.
4x 1 − 3x 2 = −2 x 1 + 4x 2 = 4 43. a. Show that if 𝐴 is invertible and 𝐴𝐵 = 𝐴𝐶, then 𝐵 = 𝐶.

If a polynomial p(x) can be factored as a product of lower degree b. Explain why part (a) and Example 3 do not contradict one
polynomials, say another.
p(x) = p1 (x)p2 (x) 44. Show that if 𝐴 is invertible and k is any nonzero scalar, then
(k𝐴)n = kn𝐴n for all integer values of n.
and if 𝐴 is a square matrix, then it can be proved that
45. a. Show that if 𝐴, 𝐵, and 𝐴 + 𝐵 are invertible matrices with
p(𝐴) = p1 (𝐴)p2 (𝐴) the same size, then
In Exercises 29–30, verify this statement for the stated matrix 𝐴 and 𝐴(𝐴−1 + 𝐵 −1 )𝐵(𝐴 + 𝐵)−1 = 𝐼
polynomials b. What does the result in part (a) tell you about the matrix
𝐴−1 + 𝐵 −1 ?
p(x) = x 2 − 9, p1 (x) = x + 3, p2 (x) = x − 3
46. A square matrix 𝐴 is said to be idempotent if 𝐴2 = 𝐴.
29. The matrix 𝐴 in Exercise 21.
a. Show that if 𝐴 is idempotent, then so is 𝐼 − 𝐴.
30. An arbitrary square matrix 𝐴. b. Show that if 𝐴 is idempotent, then 2𝐴 − 𝐼 is invertible and
is its own inverse.
31. a. Give an example of two 2 × 2 matrices such that
47. Show that if 𝐴 is a square matrix such that 𝐴k = 0 for some
(𝐴 + 𝐵)(𝐴 − 𝐵) ≠ 𝐴2 − 𝐵 2
positive integer k, then the matrix 𝐼 − 𝐴 is invertible and
b. State a valid formula for multiplying out (𝐼 − 𝐴)−1 = 𝐼 + 𝐴 + 𝐴2 + ⋅ ⋅ ⋅ + 𝐴k−1
(𝐴 + 𝐵)(𝐴 − 𝐵) 48. Show that the matrix
a b
c. What condition can you impose on 𝐴 and 𝐵 that will allow 𝐴=[ ]
you to write (𝐴 + 𝐵)(𝐴 − 𝐵) = 𝐴2 − 𝐵 2 ? c d
satisfies the equation
32. The numerical equation a2 = 1 has exactly two solutions. Find
𝐴2 − (a + d)𝐴 + (ad − bc)𝐼 = 0
at least eight solutions of the matrix equation 𝐴2 = 𝐼3 . [Hint:
Look for solutions in which all entries off the main diagonal 49. Assuming that all matrices are n × n and invertible, solve
are zero.] for 𝐷.
𝐶 𝑇 𝐵 −1 𝐴2 𝐵𝐴𝐶 −1𝐷𝐴−2 𝐵 𝑇 𝐶 −2 = 𝐶 𝑇
33. a. Show that if a square matrix 𝐴 satisfies the equation
𝐴2 + 2𝐴 + 𝐼 = 0, then 𝐴 must be invertible. What is the 50. Assuming that all matrices are n × n and invertible, solve
inverse? for 𝐷.
𝐴𝐵𝐶 𝑇 𝐷𝐵𝐴𝑇 𝐶 = 𝐴𝐵 𝑇
b. Show that if p(x) is a polynomial with a nonzero constant
term, and if 𝐴 is a square matrix for which p(𝐴) = 0, then
𝐴 is invertible. Working with Proofs
34. Is it possible for 𝐴3 to be an identity matrix without 𝐴 being In Exercises 51–58, prove the stated result.
invertible? Explain.
51. Theorem 1.4.1(a) 52. Theorem 1.4.1(b)
35. Can a matrix with a row of zeros or a column of zeros have an
inverse? Explain. 53. Theorem 1.4.1( f ) 54. Theorem 1.4.1(c)
 1.5 Elementary Matrices and a Method for Finding A−1 53

55. Theorem 1.4.2(c) 56. Theorem 1.4.2(b) Working with Technology

57. Theorem 1.4.8(d) 58. Theorem 1.4.8(e) T1. Let 𝐴 be the matrix
1 1
0 2 3⎤
⎡
True-False Exercises ⎢1 1⎥
𝐴 = ⎢4 0 5⎥
TF. In parts (a)–(k) determine whether the statement is true or ⎢1 1 ⎥
false, and justify your answer. ⎣6 7 0⎦
a. Two n × n matrices, 𝐴 and 𝐵, are inverses of one another
Discuss the behavior of 𝐴k as k increases indefinitely, that is,
if and only if 𝐴𝐵 = 𝐵 𝐴 = 0.
as k → ∞.
b. For all square matrices 𝐴 and 𝐵 of the same size, it is true
T2. In each part use your technology utility to make a conjecture
that (𝐴 + 𝐵)2 = 𝐴2 + 2𝐴𝐵 + 𝐵 2 .
about the form of 𝐴n for positive integer powers of n.
c. For all square matrices 𝐴 and 𝐵 of the same size, it is true
a 1 cos 𝜃 sin 𝜃
that 𝐴2 − 𝐵 2 = (𝐴 − 𝐵)(𝐴 + 𝐵). a. 𝐴 = [ ] b. 𝐴 = [ ]
0 a − sin 𝜃 cos 𝜃
d. If 𝐴 and 𝐵 are invertible matrices of the same size, then
𝐴𝐵 is invertible and (𝐴𝐵)−1 = 𝐴−1 𝐵 −1 . T3. The Fibonacci sequence (named for the Italian mathemati-
cian Leonardo Fibonacci 1170–1250) is
e. If 𝐴 and 𝐵 are matrices such that 𝐴𝐵 is defined, then it
is true that (𝐴𝐵)𝑇 = 𝐴𝑇 𝐵 𝑇 . 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, . . .
f. The matrix the terms of which are commonly denoted as
a b
𝐴=[ ] 𝐹0 , 𝐹1 , 𝐹2 , 𝐹3 , . . . , 𝐹 n , . . .
c d
is invertible if and only if ad − bc ≠ 0. After the initial terms 𝐹0 = 0 and 𝐹1 = 1, each term is the
g. If 𝐴 and 𝐵 are matrices of the same size and k is a con- sum of the previous two; that is,
stant, then (k𝐴 + 𝐵)𝑇 = k𝐴𝑇 + 𝐵 𝑇 .
𝐹n = 𝐹n−1 + 𝐹n−2
h. If 𝐴 is an invertible matrix, then so is 𝐴𝑇 .
Confirm that if
i. If p(x) = a0 + a1 x + a2 x 2 + ⋅ ⋅ ⋅ + am x m and 𝐼 is an iden-
tity matrix, then p(𝐼) = a0 + a1 + a2 + ⋅ ⋅ ⋅ + am . 𝐹2 𝐹1 1 1
𝑄=[ ]=[ ]
𝐹1 𝐹0 1 0
j. A square matrix containing a row or column of zeros can-
not be invertible. then
k. The sum of two invertible matrices of the same size must 𝐹n+1 𝐹n
𝑄n = [ ]
be invertible. 𝐹n 𝐹0

1.5 Elementary Matrices and a Method for

Finding A−1
In this section we will develop an algorithm for finding the inverse of a matrix, and we
will discuss some of the basic properties of invertible matrices.

Elementary Matrices
In Section 1.1 we defined three elementary row operations on a matrix 𝐴:

1. Multiply a row by a nonzero constant c.

2. Interchange two rows.
3. Add a constant c times one row to another.
60 CH APT ER 1 Systems of Linear Equations and Matrices

Exercise Set 1.5

In Exercises 1–2, determine whether the given matrix is elementary. 1 0 0 1 4

c. 𝐸 = [0 5 0], 𝐴 = [2 5]
1 0 −5 1
1. a. [ ] b. [ ] 0 0 1 3 6
−5 1 1 0
2 0 0 2 In Exercises 7–8, use the following matrices and find an elementary
1 1 0 ⎡ ⎤ matrix 𝐸 that satisfies the stated equation.
⎢0 1 0 0⎥
c. [0 0 1] d. ⎢
⎢0 0 1 0⎥⎥ 3 4 1 8 1 5
0 0 0
⎣0 0 0 1⎦ 𝐴 = [2 −7 −1], 𝐵 = [2 −7 −1]
8 1 5 3 4 1
0 0 1
1 0 3 4 1 8 1 5
2. a. [ ] b. [0 1 0]
0 √3 𝐶 = [2 −7 −1], 𝐷 = [−6 21 3]
1 0 0
2 −7 3 3 4 1
1 0 0 −1 0 0
c. [0 1 9] d. [ 0 0 1] 8 1 5
0 0 1 0 1 0 𝐹 = [8 1 1]
3 4 1
In Exercises 3–4, find a row operation and the corresponding ele-
mentary matrix that will restore the given elementary matrix to the
7. a. 𝐸𝐴 = 𝐵 b. 𝐸𝐵 = 𝐴
identity matrix.
−7 0 0 c. 𝐸𝐴 = 𝐶 d. 𝐸𝐶 = 𝐴
1 −3
3. a. [ ] b. [ 0 1 0]
0 1 8. a. 𝐸𝐵 = 𝐷 b. 𝐸𝐷 = 𝐵
0 0 1
0 0 1 0 c. 𝐸𝐵 = 𝐹 d. 𝐸𝐹 = 𝐵
1 0 0 ⎡ ⎤
⎢0 1 0 0⎥
c. [ 0 1 0] d. ⎢
⎢1 0 0 0⎥⎥
In Exercises 9–10, first use Theorem 1.4.5 and then use the inversion
−5 0 1 algorithm to find 𝐴−1 , if it exists.
⎣0 0 0 1⎦
1 4 2 −4
1 0 0 9. a. 𝐴 = [ ] b. 𝐴 = [ ]
1 0 2 7 −4 8
4. a. [ ] b. [0 1 0]
−3 1 1 −5 6 4
0 0 3 10. a. 𝐴 = [ ] b. 𝐴 = [ ]
3 −16 −3 −2
0 0 0 1 1 0 − 17 0
⎡ ⎤ ⎡ ⎤
⎢0 1 0 0⎥ ⎢0 1 0 0⎥
c. ⎢ d. ⎢ In Exercises 11–12, use the inversion algorithm to find the inverse of
⎢0 0 1 0⎥⎥ ⎢0 0 1 0⎥
⎥
the matrix (if the inverse exists).
⎣1 0 0 0⎦ ⎣0 0 0 1⎦
1 2 3 −1 3 −4
⎡ ⎤ ⎡ ⎤
⎢
11. a. 2 5 3⎥ b. ⎢ 2 4 1⎥
In Exercises 5–6 an elementary matrix 𝐸 and a matrix 𝐴 are given. ⎢ ⎥ ⎢ ⎥
Identify the row operation corresponding to 𝐸 and verify that the ⎣1 0 8⎦ ⎣−4 2 −9⎦
product 𝐸𝐴 results from applying the row operation to 𝐴. 1 1
− 25 1 1
− 25
0 1 −1 −2 5 −1 ⎡5 5 ⎤ ⎡5 5 ⎤
5. a. 𝐸 = [ ], 𝐴=[ ] ⎢1 1 1 ⎥ ⎢2 − 35 3 ⎥
− 10
1 0 3 −6 −6 −6 12. a. ⎢ 5 5 10 ⎥ b. ⎢ 5 ⎥
⎢1 1 ⎥ ⎢1 1 ⎥
⎣5 − 45 10 ⎦ ⎣5 − 45 10 ⎦
1 0 0 2 −1 0 −4 −4
b. 𝐸 = [0 1 0], 𝐴 = [1 −3 −1 5 3]
0 −3 1 2 0 1 3 −1 In Exercises 13–18, use the inversion algorithm to find the inverse of
the matrix (if the inverse exists).
1 0 4 1 4 1 0 1 √ 3 √2
⎡ 2 0⎤
c. 𝐸 = [0 1 0], 𝐴 = [2 5] 14. ⎢−4√2
13. [0 1 1]
⎢
√2 0⎥
⎥
0 0 1 3 6 1 1 0 ⎣ 0 0 1⎦
−6 0 −1 −2 5 −1
6. a. 𝐸 = [ ], 𝐴=[ ] 1 0 0 0
0 1 3 −6 −6 −6 2 6 6 ⎡ ⎤
⎢1 3 0 0⎥
15. [2 7 6] 16. ⎢
1 0 0 2 −1 0 −4 −4
⎢1 3 5 0⎥⎥
2 7 7
b. 𝐸 = [−4 1 0], 𝐴 = [1 −3 −1 5 3] ⎣1 3 5 7⎦
0 0 1 2 0 1 3 −1
 1.5 Elementary Matrices and a Method for Finding A−1 61

2 −4 0 0 0 0 2 0 Working with Proofs

⎡ ⎤ ⎡ ⎤
⎢1 2 12 0⎥ ⎢1 0 0 1⎥
17. ⎢ 18. ⎢ 31. Prove that if 𝐴 and 𝐵 are m × n matrices, then 𝐴 and 𝐵 are
⎢0 0 2 0⎥⎥ ⎢0 −1 3 0⎥⎥ row equivalent if and only if 𝐴 and 𝐵 have the same reduced
⎣0 −1 −4 −5⎦ ⎣2 1 5 −3⎦ row echelon form.

In Exercises 19–20, find the inverse of each of the following 4 × 4 32. Prove that if 𝐴 is an invertible matrix and 𝐵 is row equivalent
matrices, where k1 , k2 , k3 , k4 , and k are all nonzero. to 𝐴, then 𝐵 is also invertible.

k 0 0 0 k 1 0 0 33. Prove that if 𝐵 is obtained from 𝐴 by performing a sequence

⎡ 1 ⎤ ⎡ ⎤
⎢0 k2 0 0⎥ ⎢0 1 0 0⎥ of elementary row operations, then there is a second sequence
19. a. ⎢ b. ⎢
⎢0 0 k3 0⎥ ⎥ ⎢0 0 k 1⎥⎥
of elementary row operations, which when applied to 𝐵 recov-
ers 𝐴.
⎣0 0 0 k4 ⎦ ⎣0 0 0 1⎦

0 0 0 k1 k 0 0 0
⎡ ⎤ ⎡ ⎤ True-False Exercises
⎢0 0 k2 0⎥ ⎢1 k 0 0⎥
20. a. ⎢ b. ⎢
⎢0 k3 0 0⎥ ⎥ ⎢0 1 k 0⎥⎥
TF. In parts (a)–(g) determine whether the statement is true or
false, and justify your answer.
⎣k4 0 0 0⎦ ⎣0 0 1 k⎦
a. The product of two elementary matrices of the same size
In Exercises 21–22, find all values of c, if any, for which the given must be an elementary matrix.
matrix is invertible.
c c c c 1 0 b. Every elementary matrix is invertible.
21. [1 c c] 22. [1 c 1] c. If 𝐴 and 𝐵 are row equivalent, and if 𝐵 and 𝐶 are row
1 1 c 0 1 c equivalent, then 𝐴 and 𝐶 are row equivalent.
In Exercises 23–26, express the matrix and its inverse as products of d. If 𝐴 is an n × n matrix that is not invertible, then the lin-
elementary matrices. ear system 𝐴x = 0 has infinitely many solutions.
−3 1 1 0
23. [ ] 24. [ ] e. If 𝐴 is an n × n matrix that is not invertible, then the
2 2 −5 2
matrix obtained by interchanging two rows of 𝐴 cannot
1 0 −2 1 1 0 be invertible.
25. [0 4 3] 26. [1 1 1]
f. If 𝐴 is invertible and a multiple of the first row of 𝐴
0 0 1 0 1 1 is added to the second row, then the resulting matrix is
invertible.
In Exercises 27–28, show that the matrices 𝐴 and 𝐵 are row equiv-
alent by finding a sequence of elementary row operations that pro- g. An expression of an invertible matrix 𝐴 as a product of
duces 𝐵 from 𝐴, and then use that result to find a matrix 𝐶 such elementary matrices is unique.
that 𝐶𝐴 = 𝐵.
1 2 3 1 0 5
27. 𝐴 = [1 4 1], 𝐵 = [0 2 −2] Working with Technology
2 1 9 1 1 4 T1. It can be proved that if the partitioned matrix
2 1 0 6 9 4
28. 𝐴 = [−1 1 0], 𝐵 = [−5 −1 0] 𝐴 𝐵
[ ]
3 0 −1 −1 −2 −1 𝐶 𝐷

29. Show that if is invertible, then its inverse is

1 0 0
𝐴 = [ 0 1 0]
𝐴−1 + 𝐴−1 𝐵(𝐷 − 𝐶𝐴−1 𝐵)−1 𝐶𝐴−1 −𝐴−1 𝐵(𝐷 − 𝐶𝐴−1 𝐵)−1
a b c [ ]
is an elementary matrix, then at least one entry in the third −(𝐷 − 𝐶𝐴−1 𝐵)−1 𝐶𝐴−1 (𝐷 − 𝐶𝐴−1 𝐵)−1
row must be zero.
30. Show that provided that all of the inverses on the right side exist. Use
0 a 0 0 0 this result to find the inverse of the matrix
⎡ ⎤
⎢b 0 c 0 0⎥
⎢ ⎥
𝐴 = ⎢0 d 0 e 0⎥ 1 2 1 0
⎢0 ⎡ ⎤
0 𝑓 0 g⎥ ⎢0 −1 0 1⎥
⎢ ⎥ ⎢ ⎥
⎣0 0 0 h 0⎦ ⎢0 0 2 0⎥
⎢ ⎥
is not invertible for any values of the entries. ⎣0 0 3 3⎦
 1.6 More on Linear Systems and Invertible Matrices 67

It is now evident from the third row in the matrix that the system has a solution if and only
if b1 , b2 , and b3 satisfy the condition
b3 − b2 − b1 = 0 or b3 = b1 + b2
To express this condition another way, 𝐴x = b is consistent if and only if b is a matrix of the
form
b1
b = [ b2 ]
b1 + b2
where b1 and b2 are arbitrary.

EXAMPLE 4 | Determining Consistency by Elimination

What conditions must b1 , b2 , and b3 satisfy in order for the system of equations
x 1 + 2x 2 + 3x 3 = b1
2x 1 + 5x 2 + 3x 3 = b2
x1 + 8x 3 = b3
to be consistent?
Solution The augmented matrix is
1 2 3 b1
[2 5 3 b2 ]
1 0 8 b3
Reducing this to reduced row echelon form yields (verify)
1 0 0 −40b1 + 16b2 + 9b3
[0 1 0 13b1 − 5b2 − 3b3 ] (2)
0 0 1 5b1 − 2b2 − b3
In this case there are no restrictions on b1 , b2 , and b3 , so the system has the unique solution
What does the result in
x 1 = −40b1 + 16b2 + 9b3 , x 2 = 13b1 − 5b2 − 3b3 , x 3 = 5b1 − 2b2 − b3 (3)
Example 4 tell you about
for all values of b1 , b2 , and b3 . the coefficient matrix of the
system?

Exercise Set 1.6

In Exercises 1–8, solve the system by inverting the coefficient matrix 7. 3x 1 + 5x 2 = b1 8. x 1 + 2x 2 + 3x 3 = b1

and using Theorem 1.6.2. x 1 + 2x 2 = b2 2x 1 + 5x 2 + 5x 3 = b2
3x 1 + 5x 2 + 8x 3 = b3
1. x 1 + x 2 = 2 2. 4x 1 − 3x 2 = −3
5x 1 + 6x 2 = 9 2x 1 − 5x 2 = 9 In Exercises 9–12, solve the linear systems. Using the given values for
the b’s solve the systems together by reducing an appropriate aug-
mented matrix to reduced row echelon form.
3. x 1 + 3x 2 + x 3 = 4 4. 5x 1 + 3x 2 + 2x 3 = 4
2x 1 + 2x 2 + x 3 = −1 3x 1 + 3x 2 + 2x 3 = 2 9. x 1 − 5x 2 = b1
2x 1 + 3x 2 + x 3 = 3 x 2 + x3 = 5 3x 1 + 2x 2 = b2
i. b1 = 1, b2 = 4 ii. b1 = −2, b2 = 5

5. x+y+ z= 5 6. − x − 2y − 3z = 0 10. −x 1 + 4x 2 + x 3 = b1
x + y − 4z = 10 𝑤 + x + 4y + 4z = 7 x 1 + 9x 2 − 2x 3 = b2
−4x + y + z = 0 𝑤 + 3x + 7 y + 9z = 4 6x 1 + 4x 2 − 8x 3 = b3
−𝑤 − 2x − 4y − 6z = 6 i. b1 = 0, b2 = 1, b3 = 0 ii. b1 = −3, b2 = 4, b3 = −5
68 CH APT ER 1 Systems of Linear Equations and Matrices

11. 4x 1 − 7x 2 = b1 tem can be written in the form x = x1 + x0 , where x0 is a solu-

x 1 + 2x 2 = b2 tion to 𝐴x = 0. Prove also that every matrix of this form is a
i. b1 = 0, b2 = 1 ii. b1 = −4, b2 = 6 solution.
iii. b1 = −1, b2 = 3 iv. b1 = −5, b2 = 1
24. Use part (a) of Theorem 1.6.3 to prove part (b).
12. x 1 + 3x 2 + 5x 3 = b1
−x 1 − 2x 2 = b2 True-False Exercises
2x 1 + 5x 2 + 4x 3 = b3
TF. In parts (a)–(g) determine whether the statement is true or
i. b1 = 1, b2 = 0, b3 = −1
false, and justify your answer.
ii. b1 = 0, b2 = 1, b3 = 1
iii. b1 = −1, b2 = −1, b3 = 0 a. It is impossible for a system of linear equations to have
exactly two solutions.
In Exercises 13–17, determine conditions on the bi ’s, if any, in order
to guarantee that the linear system is consistent. b. If 𝐴 is a square matrix, and if the linear system 𝐴x = b
has a unique solution, then the linear system 𝐴x = c also
13. x 1 + 3x 2 = b1 14. 6x 1 − 4x 2 = b1
must have a unique solution.
−2x 1 + x 2 = b2 3x 1 − 2x 2 = b2

15. x 1 − 2x 2 + 5x 3 = b1 16. x 1 − 2x 2 − x 3 = b1 c. If 𝐴 and 𝐵 are n × n matrices such that 𝐴𝐵 = 𝐼n , then

4x 1 − 5x 2 + 8x 3 = b2 −4x 1 + 5x 2 + 2x 3 = b2 𝐵𝐴 = 𝐼n .
−3x 1 + 3x 2 − 3x 3 = b3 −4x 1 + 7x 2 + 4x 3 = b3
d. If 𝐴 and 𝐵 are row equivalent matrices, then the linear
17. x1 − x2 + 3x 3 + 2x 4 = b1 systems 𝐴x = 0 and 𝐵x = 0 have the same solution set.
−2x 1 + x2 + 5x 3 + x4 = b2
−3x 1 + 2x 2 + 2x 3 − x4 = b3 e. Let 𝐴 be an n × n matrix and 𝑆 is an n × n invertible
4x 1 − 3x 2 + x3 + 3x 4 = b4 matrix. If x is a solution to the system (𝑆 −1 𝐴𝑆)x = b,
then 𝑆x is a solution to the system 𝐴y = 𝑆b.
18. Consider the matrices
2 1 2 x1 f. Let 𝐴 be an n × n matrix. The linear system 𝐴x = 4x has
𝐴 = [2 2 −2] and x = [x 2 ] a unique solution if and only if 𝐴 − 4𝐼 is an invertible
matrix.
3 1 1 x3
a. Show that the equation 𝐴x = x can be rewritten as g. Let 𝐴 and 𝐵 be n × n matrices. If 𝐴 or 𝐵 (or both) are not
(𝐴 − 𝐼)x = 0 and use this result to solve 𝐴x = x for x. invertible, then neither is 𝐴𝐵.
b. Solve 𝐴x = 4x. Working with Technology
In Exercises 19–20, solve the matrix equation for 𝑋. T1. Colors in print media, on computer monitors, and on tele-
vision screens are implemented using what are called “color
1 −1 1 2 −1 5 7 8 models.” For example, in the RGB model, colors are created by
19. [2 3 0] 𝑋 = [4 0 −3 0 1] mixing percentages of red (R), green (G), and blue (B), and in
0 2 −1 3 5 −7 2 1 the YIQ model (used in TV broadcasting), colors are created
by mixing percentages of luminescence (Y) with percentages
of a chrominance factor (I) and a chrominance factor (Q). The
−2 0 1 4 3 2 1 conversion from the RGB model to the YIQ model is accom-
20. [ 0 −1 −1] 𝑋 = [6 7 8 9] plished by the matrix equation
1 1 −4 1 3 7 9
Y .299 .587 .114 R
⎡ ⎤ ⎡ ⎤⎡ ⎤
⎢ I ⎥ = ⎢.596 −.275 −.321⎥ ⎢G⎥
Working with Proofs ⎢ ⎥ ⎢ ⎥⎢ ⎥
⎣Q⎦ ⎣.212 −.523 .311⎦ ⎣ B ⎦
21. Let 𝐴x = 0 be a homogeneous system of n linear equations in
n unknowns that has only the trivial solution. Prove that if k What matrix would you use to convert the YIQ model to the
is any positive integer, then the system 𝐴k x = 0 also has only RGB model?
the trivial solution. T2. Let
22. Let 𝐴x = 0 be a homogeneous system of n linear equations 1 −2 2 0 11 1
in n unknowns, and let 𝑄 be an invertible n × n matrix. ⎡ ⎤ ⎡ ⎤ ⎡ ⎤ ⎡ ⎤
𝐴 = ⎢4 5 1⎥, 𝐵1 = ⎢1⎥, 𝐵2 = ⎢ 5⎥, 𝐵3 = ⎢−4⎥
Prove that 𝐴x = 0 has only the trivial solution if and only if ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
(𝑄𝐴)x = 0 has only the trivial solution. ⎣0 3 −1⎦ ⎣ 7⎦ ⎣ 3⎦ ⎣ 2⎦

23. Let 𝐴x = b be any consistent system of linear equations, and Solve the linear systems 𝐴x = 𝐵1 , 𝐴x = 𝐵2 , 𝐴x = 𝐵3 using
let x1 be a fixed solution. Prove that every solution to the sys- the method of Example 2.
74 CH APT ER 1 Systems of Linear Equations and Matrices

EXAMPLE 6 | The Product of a Matrix and Its Transpose

Is Symmetric

Let 𝐴 be the 2 × 3 matrix

1 −2 4
𝐴=[ ]
3 0 −5
Then
1 3 10 −2 −11
1 −2 4
𝐴𝑇𝐴 = [−2 0] [ ] = [ −2 4 −8]
3 0 −5
4 −5 −11 −8 41
1 3
𝑇
1 −2 4 21 −17
𝐴𝐴 = [ ][ −2 0] = [ ]
3 0 −5 −17 34
4 −5
Observe that 𝐴𝑇𝐴 and 𝐴𝐴𝑇 are symmetric as expected.

Exercise Set 1.7

In Exercises 1–2, classify the matrix as upper triangular, lower tri- 2 0 0 4 −1 3 −3 0 0

angular, or diagonal, and decide by inspection whether the matrix 6. [0 −1 0] [ 1 2 0] [ 0 5 0]
is invertible. Recall that a diagonal matrix is both upper and lower
0 0 4 −5 1 −2 0 0 2
triangular, so there may be more than one answer in some parts.
In Exercises 7–10, find 𝐴2 , 𝐴−2 , and 𝐴−k (where k is any integer) by
2 1 0 0 inspection.
1. a. [ ] b. [ ]
0 3 4 0
1 0 −6 0 0
7. 𝐴 = [ ]
−1 0 0 3 −2 7 0 −2 8. 𝐴 = [ 0 3 0]
⎡ ⎤ ⎡ ⎤
c. ⎢ 0 2 0⎥ d. ⎢0 0 3⎥ 0 0 5
⎢ ⎢ ⎥
1⎥
⎣ 0 0 5⎦ ⎣0 0 8⎦ 1
0 0 −2 0 0 0
⎡2 ⎤ ⎡ ⎤
1
9. 𝐴 = ⎢ 0 3 0⎥ ⎢ 0 −4 0 0⎥
⎢ 10. 𝐴 = ⎢
4 0 0 −3 1⎥ 0 0 −3 0⎥
2. a. [ ] b. [ ] ⎣0 0 4⎦ ⎢ ⎥
1 7 0 0 ⎣ 0 0 0 2⎦
4 0 0 3 0 0 In Exercises 11–12, compute the product by inspection.
⎡ ⎤ ⎡ ⎤
3
c. ⎢0 0⎥ d. ⎢3 1 0⎥
⎢ 5 ⎥ ⎢ ⎥ 1 0 0 2 0 0 0 0 0
⎣7 0 0⎦ ⎡ ⎤⎡ ⎤⎡ ⎤
⎣0 0 −2⎦ ⎢
11. 0 0 ⎥
0 0⎢ 5 ⎥
0 0⎢ 2 0⎥
⎢ ⎥⎢ ⎥⎢ ⎥
In Exercises 3–6, find the product by inspection. ⎣0 0 3⎦ ⎣ 0 0 0⎦ ⎣0 0 1⎦

3 0 0 2 1 −1 0 0 3 0 0 5 0 0
3. [0 −1 0] [−4 1] ⎡ ⎤⎡ ⎤⎡ ⎤
12. ⎢ 0 2 0⎥ ⎢0 5 0⎥ ⎢ 0 −2 0⎥
0 0 2 2 5 ⎢ ⎥⎢ ⎥⎢ ⎥
⎣ 0 0 4⎦ ⎣0 0 7⎦ ⎣ 0 0 3⎦
−4 0 0 In Exercises 13–14, compute the indicated quantity.
1 2 −5
4. [ ][ 0 3 0]
−3 −1 0 1 0
39
1 0
1000
0 0 2
13. [ ] 14. [ ]
0 −1 0 −1
5 0 0 −3 2 0 4 −4
In Exercises 15–16, use what you have learned in this section
5. [0 2 0] [ 1 −5 3 0 3]
about multiplying by diagonal matrices to compute the product by
0 0 −3 −6 2 2 2 2
inspection.
 1.7 Diagonal, Triangular, and Symmetric Matrices 75

a 0 0 u 𝑣 r s ta 0 0 29. If 𝐴 is an invertible upper triangular or lower triangular

⎡ ⎤⎡ ⎤ ⎡ ⎤⎡ ⎤
15. a. ⎢ 0 b 0⎥ ⎢ 𝑤 x⎥ b. ⎢u 𝑣 𝑤⎥ ⎢0 b 0⎥ matrix, what can you say about the diagonal entries of 𝐴−1 ?
⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥
⎣0 0 c⎦ ⎣ y z⎦ ⎣x y z⎦ ⎣ 0 0 c⎦ 30. Show that if 𝐴 is a symmetric n × n matrix and 𝐵 is any n × m
matrix, then the following products are symmetric:
u 𝑣 a 0 0 r s t
⎡ ⎤ a 0 ⎡ ⎤⎡ ⎤
⎢
16. a. 𝑤 x⎥ [ ] ⎢
b. 0 b ⎥
0 u⎢ 𝑣 𝑤⎥ 𝐵 𝑇 𝐵, 𝐵𝐵 𝑇 , 𝐵 𝑇𝐴𝐵
⎢ ⎥ 0 b ⎢ ⎥⎢ ⎥
⎣ y z⎦ ⎣0 0 c⎦ ⎣ x y z⎦
In Exercises 31–32, find a diagonal matrix 𝐴 that satisfies the given
In Exercises 17–18, create a symmetric matrix by substituting appro- condition.
priate numbers for the ×’s. 1 0 0 9 0 0
1 × × × 31. 𝐴5 = [0 −1 0] 32. 𝐴−2 = [0 4 0]
⎡ ⎤
2 −1 ⎢3 1 × ×⎥ 0 0 −1 0 0 1
17. a. [ ] b. ⎢ ⎥
× 3 ⎢7 −8 0 ×⎥ 33. Verify Theorem 1.7.1(b) for the matrix product 𝐴𝐵 and Theo-
⎢ ⎥
⎣2 −3 9 0⎦ rem 1.7.1(d) for the matrix 𝐴, where

1 7 −3 2 −1 2 5 2 −8 0
⎡ ⎤
0 × ⎢× 4 5 −7⎥ 𝐴=[ 0 1 3], 𝐵 = [0 2 1]
18. a. [ ] b. ⎢ ⎥ 0 0 −4 0 0 3
3 0 ⎢× × 1 −6⎥
⎢ ⎥
⎣× × × 3⎦ 34. Let 𝐴 be an n × n symmetric matrix.

In Exercises 19–22, determine by inspection whether the matrix is a. Show that 𝐴2 is symmetric.
invertible. b. Show that 2𝐴2 − 3𝐴 + 𝐼 is symmetric.
0 6 −1 −1 2 4
35. Verify Theorem 1.7.4 for the given matrix 𝐴.
19. [0 7 −4] 20. [ 0 3 0]
0 0 −2 0 0 5 1 −2 3
2 −1
1 0 0 0 2 0 0 0 a. 𝐴 = [ ] b. 𝐴 = [−2 1 −7]
⎡ ⎤ ⎡ ⎤ −1 3
3 −7 4
⎢2 −5 0 0⎥ ⎢−3 −1 0 0⎥
21. ⎢ 22. ⎢
⎢4 −3 4 0⎥⎥ ⎢−4 −6 0 0⎥⎥ 36. Find all 3 × 3 diagonal matrices 𝐴 that satisfy
⎣1 −2 1 3⎦ ⎣ 0 3 8 −5⎦ 𝐴2 − 3𝐴 − 4𝐼 = 0.

37. Let 𝐴 = [aij ] be an n × n matrix. Determine whether 𝐴 is sym-

In Exercises 23–24, find the diagonal entries of 𝐴𝐵 by inspection. metric.
3 2 6 −1 2 7 a. aij = i2 + j2 b. aij = i2 − j2
23. 𝐴 = [0 1 −2], 𝐵=[ 0 5 3] c. aij = 2i + 2j d. aij = 2i2 + 2j3
0 0 −1 0 0 6
38. On the basis of your experience with Exercise 37, devise a gen-
4 0 0 6 0 0 eral test that can be applied to a formula for aij to determine
24. 𝐴 = [−2 0 0], 𝐵 = [1 5 0] whether 𝐴 = [aij ] is symmetric.
−3 0 7 3 2 6 39. Find an upper triangular matrix that satisfies
In Exercises 25–26, find all values of the unknown constant(s) for 1 30
which 𝐴 is symmetric. 𝐴3 = [ ]
0 −8
4 −3
25. 𝐴 = [ ] 40. If the n × n matrix 𝐴 can be expressed as 𝐴 = 𝐿𝑈, where 𝐿 is
a+5 −1 a lower triangular matrix and 𝑈 is an upper triangular matrix,
then the linear system 𝐴x = b can be expressed as 𝐿𝑈x = b
2 a − 2b + 2c 2a + b + c and can be solved in two steps:
26. 𝐴 = [3 5 a+c ]
Step 1. Let 𝑈x = y, so that 𝐿𝑈x = b can be expressed as
0 −2 7 𝐿y = b. Solve this system.
In Exercises 27–28, find all values of x for which 𝐴 is invertible. Step 2. Solve the system 𝑈x = y for x.
In each part, use this two-step method to solve the given
x−1 x2 x4 system.
27. 𝐴 = [ 0 x+2 x3 ] 1 0 0 2 −1 3 x1 1
0 0 x−4 a. [−2 3 0] [ 0 1 2] [x 2 ] = [−2]
2 4 1 0 0 4 x3 0
x − 12 0 0
⎡ ⎤
⎢ 2 0 0 3 −5 2 x1 4
28. 𝐴 = ⎢ x x− 1
3 0 ⎥⎥
⎢ 2 b. [ 4 1 0] [0 4 1] [x 2 ] = [−5]
1⎥
⎣ x x3 x+ 4⎦ −3 −2 3 0 0 2 x3 2
76 CH APT ER 1 Systems of Linear Equations and Matrices

In the text we defined a matrix 𝐴 to be symmetric if 𝐴𝑇 = 𝐴. Anal- c. The sum of an upper triangular matrix and a lower trian-
ogously, a matrix 𝐴 is said to be skew-symmetric if 𝐴𝑇 = −𝐴. gular matrix is a diagonal matrix.
Exercises 41–45 are concerned with matrices of this type.
d. All entries of a symmetric matrix are determined by the
41. Fill in the missing entries (marked with ×) so the matrix 𝐴 is
entries occurring on and above the main diagonal.
skew-symmetric.
× × 4 × 0 × e. All entries of an upper triangular matrix are determined
a. 𝐴 = [ 0 × ×] b. 𝐴 = [× × −4] by the entries occurring on and above the main diagonal.
× −1 × 8 × × f. The inverse of an invertible lower triangular matrix is an
42. Find all values of a, b, c, and d for which 𝐴 is skew-symmetric. upper triangular matrix.
0 2a − 3b + c 3a − 5b + 5c g. A diagonal matrix is invertible if and only if all of its diag-
𝐴 = [−2 0 5a − 8b + 6c] onal entries are positive.
−3 −5 d
h. The sum of a diagonal matrix and a lower triangular
43. We showed in the text that the product of symmetric matrices matrix is a lower triangular matrix.
is symmetric if and only if the matrices commute. Is the prod-
uct of commuting skew-symmetric matrices skew-symmetric? i. A matrix that is both symmetric and upper triangular
Explain. must be a diagonal matrix.

j. If 𝐴 and 𝐵 are n × n matrices such that 𝐴 + 𝐵 is symmet-

Working with Proofs ric, then 𝐴 and 𝐵 are symmetric.

44. Prove that every square matrix 𝐴 can be expressed as the sum k. If 𝐴 and 𝐵 are n × n matrices such that 𝐴 + 𝐵 is upper
of a symmetric matrix and a skew-symmetric matrix. [Hint: triangular, then 𝐴 and 𝐵 are upper triangular.
Note the identity 𝐴 = 12 (𝐴 + 𝐴𝑇 ) + 12 (𝐴 − 𝐴𝑇 ).]
l. If 𝐴2 is a symmetric matrix, then 𝐴 is a symmetric matrix.
45. Prove the following facts about skew-symmetric matrices.
m. If k𝐴 is a symmetric matrix for some k ≠ 0, then 𝐴 is a
a. If 𝐴 is an invertible skew-symmetric matrix, then 𝐴−1 is symmetric matrix.
skew-symmetric.
b. If 𝐴 and 𝐵 are skew-symmetric matrices, then so are 𝐴𝑇 ,
𝐴 + 𝐵, 𝐴 − 𝐵, and k𝐴 for any scalar k. Working with Technology
46. Prove: If the matrices 𝐴 and 𝐵 are both upper triangular or T1. Starting with the formula stated in Exercise T1 of Section 1.5,
both lower triangular, then the diagonal entries of both 𝐴𝐵 derive a formula for the inverse of the “block diagonal” matrix
and 𝐵𝐴 are the products of the diagonal entries of 𝐴 and 𝐵. 𝐷1 0
𝑇 2
[ ]
47. Prove: If 𝐴 𝐴 = 𝐴, then 𝐴 is symmetric and 𝐴 = 𝐴 . 0 𝐷2
in which 𝐷1 and 𝐷2 are invertible, and use your result to com-
True-False Exercises pute the inverse of the matrix
TF. In parts (a)–(m) determine whether the statement is true or 1.24 2.37 0 0
false, and justify your answer. ⎡ ⎤
⎢3.08 −1.01 0 0 ⎥
a. The transpose of a diagonal matrix is a diagonal matrix. 𝑀=⎢ ⎥
⎢ 0 0 2.76 4.92⎥
⎢ ⎥
b. The transpose of an upper triangular matrix is an upper ⎣ 0 0 3.23 5.54⎦
triangular matrix.

1.8 Introduction to Linear Transformations

Up to now we have treated matrices simply as rectangular arrays of numbers and have
been concerned primarily with developing algebraic properties of those arrays. In this
section we will view matrices in a completely different way. Here we will be concerned
with how matrices can be used to transform or “map” one vector into another by matrix
multiplication. This will be the foundation for much of our work in subsequent sections.

Recall that in Section 1.1 we defined an “ordered n-tuple” to be a sequence of n real num-
bers, and we observed that a solution of a linear system in n unknowns, say
x 1 = s1 , x 2 = s2 , . . . , x n = sn
124 C H APT ER 2 Determinants

In the 2 × 2 case, the determinant can be computed by forming the product of the entries
on the rightward arrow and subtracting the product of the entries on the leftward arrow.
In the 3 × 3 case we first recopy the first and second columns as shown in the figure, after
which we can compute the determinant by summing the products of the entries on the
Warning The arrow rightward arrows and subtracting the products on the leftward arrows. These procedures
technique works only for execute the computations
determinants of 2 × 2 and
| a11 a12 |
3 × 3 matrices. It does not | | = a11 a22 − a12 a21
| a21 a22 |
work for matrices of size
4 × 4 or higher. | a11 a12 a13 |
| | |a a | |a a | |a a |
| a21 a22 a23 | = a11 | 22 23 | − a12 | 21 23 | + a13 | 21 22 |
| | | a32 a33 | | a31 a33 | | a31 a32 |
| a31 a32 a33 |
= a11 (a22 a33 − a23 a32 ) − a12 (a21 a33 − a23 a31 ) + a13 (a21 a32 − a22 a31 )
= a11 a22 a33 + a12 a23 a31 + a13 a21 a32 − a13 a22 a31 − a12 a21 a33 − a11 a23 a32
which agrees with the cofactor expansions along the first row.

EXAMPLE 7 | A Technique for Evaluating 2 × 2 and

3 × 3 Determinants

3 1 3 1
= = (3)( 2) (1)(4) = 10
4 2 4 2

1 2 3 1 2 3 1 2
4 5 6 = 4 5 6 4 5
7 8 9 7 8 9 7 8

= [45 + 84 + 96] [105 48 72] = 240

Exercise Set 2.1

In Exercises 1–2, find all the minors and cofactors of the matrix 𝐴. 4. Let
1 −2 3 1 1 2 2 3 −1 1
⎡ ⎤
1. 𝐴 = [ 6 7 −1] 2. 𝐴 = [3 3 6] ⎢−3 2 0 3⎥
𝐴=⎢
−3 1 4 0 1 4
⎢ 3 −2 1 0⎥⎥
⎣ 3 −2 1 4⎦
3. Let
Find
4 −1 1 6
⎡ ⎤ a. 𝑀32 and 𝐶32 . b. 𝑀44 and 𝐶44 .
⎢0 0 −3 3⎥
𝐴=⎢
⎢4 1 0 14⎥⎥ c. 𝑀41 and 𝐶41 . d. 𝑀24 and 𝐶24 .
⎣4 1 3 2⎦
In Exercises 5–8, evaluate the determinant of the given matrix. If the
Find matrix is invertible, use Equation (2) to find its inverse.
a. 𝑀13 and 𝐶13 . b. 𝑀23 and 𝐶23 . 3 5 4 1 −5 7 √2 √6
5. [ ] 6. [ ] 7. [ ] 8. [ ]
c. 𝑀22 and 𝐶22 . d. 𝑀21 and 𝐶21 . −2 4 8 2 −7 −2 4 √3
 2.1 Determinants by Cofactor Expansion 125

In Exercises 9–14, use the arrow technique of Figure 2.1.1 to evaluate In Exercises 27–32, evaluate the determinant of the given matrix by
the determinant. inspection.

| −2 7 6| 1 0 0 2 0 0
|a − 3 5 | | | 27. [0 −1 0] 28. [0 2 0]
9. | | 10. | 5 1 −2 |
| −3 a − 2| | | 0 0 1 0 0 2
| 3 8 4|
0 0 0 0 1 1 1 1
| −2 1 4| | −1 1 2| ⎡ ⎤ ⎡ ⎤
| | | | ⎢1 2 0 0⎥ ⎢0 2 2 2⎥
11. | 3 5 −7 | 12. | 3 0 −5 | 29. ⎢ 30. ⎢
| | | | ⎢0 4 3 0⎥⎥ ⎢0 0 3 3⎥⎥
| 1 6 2| | 1 7 2|
⎣1 2 3 8⎦ ⎣0 0 0 4⎦
|3 0 0| |c −4 3| 1 2 7 −3 −3 0 0 0
| | | | ⎡ ⎤ ⎡ ⎤
13. | 2 −1 5| 14. | 2 1 c2 | ⎢0 1 −4 1⎥ ⎢ 1 2 0 0⎥
| | | | 31. ⎢ 32. ⎢
|1 9 −4 | |4 c−1 2|
⎢0 0 2 7⎥⎥ ⎢ 40 10 −1 0⎥⎥
In Exercises 15–18, find all values of 𝜆 for which det(𝐴) = 0. ⎣0 0 0 3⎦ ⎣100 200 −23 3⎦
33. In each part, show that the value of the determinant is inde-
𝜆−4 0 0
𝜆−2 1 pendent of 𝜃.
15. 𝐴 = [ ] 16. 𝐴 = [ 0 𝜆 2 ]
−5 𝜆+4 | sin 𝜃 cos 𝜃 |
0 3 𝜆−1 a. | |
| − cos 𝜃 sin 𝜃 |
𝜆−4 4 0
𝜆−1 0 | sin 𝜃 cos 𝜃 0|
17. 𝐴 = [ ] 18. 𝐴 = [ −1 𝜆 0 ] | |
2 𝜆+1 b. | − cos 𝜃 sin 𝜃 0|
0 0 𝜆−5 | |
| sin 𝜃 − cos 𝜃 sin 𝜃 + cos 𝜃 1|
19. Evaluate the determinant in Exercise 13 by a cofactor expan- 34. Show that the matrices
sion along
a b d e
a. the first row. b. the first column. 𝐴=[ ] and 𝐵=[ ]
0 c 0 𝑓
c. the second row. d. the second column.
commute if and only if
e. the third row. f. the third column.
|b a−c |
| |=0
20. Evaluate the determinant in Exercise 12 by a cofactor expan- |e d−𝑓|
sion along
35. By inspection, what is the relationship between the following
a. the first row. b. the first column. determinants?
c. the second row. d. the second column. |a b c| |a + 𝜆 b c|
| | | |
e. the third row. f. the third column. d1 = | d 1 𝑓| and d2 = | d 1 𝑓|
| | | |
|g 0 1| | g 0 1|
In Exercises 21–26, evaluate det(𝐴) by a cofactor expansion along a
row or column of your choice. 36. Show that
−3 0 7 3 3 1 1 | tr(𝐴) 1 |
det(𝐴) = | |
21. 𝐴 = [ 2 5 1] 22. 𝐴 = [1 0 −4] 2 | tr(𝐴2 ) tr(𝐴) |
−1 0 5 1 −3 5 for every 2 × 2 matrix 𝐴.

1 k k2 k+1 k−1 7 37. What can you say about an nth-order determinant all of whose
entries are 1? Explain.
23. 𝐴 = [1 k k2 ] 24. 𝐴 = [ 2 k−3 4]
1 k k2 5 k+1 k 38. What is the maximum number of zeros that a 3 × 3 matrix can
have without having a zero determinant? Explain.
3 3 0 5 39. Explain why the determinant of a matrix with integer entries
⎡ ⎤
⎢2 2 0 −2⎥ must be an integer.
25. 𝐴 = ⎢
⎢4 1 −3 0⎥⎥
⎣2 10 3 2⎦ Working with Proofs
4 0 0 1 0 40. Prove that (x 1 , y1 ), (x 2 , y2 ), and (x 3 , y3 ) are collinear points if
⎡ ⎤ and only if
⎢3 3 3 −1 0⎥
⎢ ⎥
26. 𝐴 = ⎢1 2 4 2 3⎥ | x1 y1 1|
⎢9 | |
⎢ 4 6 2 3⎥⎥ | x2 y2 1| = 0
| |
⎣2 2 4 2 3⎦ | x3 y3 1|
126 C H APT ER 2 Determinants

41. Prove that the equation of the line through the distinct points c. The minor 𝑀ij is the same as the cofactor 𝐶ij if i + j is
(a1 , b1 ) and (a2 , b2 ) can be written as even.

|x y 1| d. If 𝐴 is a 3 × 3 symmetric matrix, then 𝐶ij = 𝐶ji for all i

| | and j.
| a1 b1 1| = 0
| |
| a2 b2 1| e. The number obtained by a cofactor expansion of a matrix
𝐴 is independent of the row or column chosen for the
42. Prove that if 𝐴 is upper triangular and 𝐵ij is the matrix that
expansion.
results when the ith row and jth column of 𝐴 are deleted, then
𝐵ij is upper triangular if i < j. f. If 𝐴 is a square matrix whose minors are all zero, then
det(𝐴) = 0.
43. A matrix in which the entries in each row (or in each col-
umn) form a geometric progression starting with 1 is called g. The determinant of a lower triangular matrix is the sum
a Vandermonde matrix in honor of the French medical of the entries along the main diagonal.
doctor, mathematician, and musician Alexandre-Théophile
Vandermonde (February 28, 1735–January 1, 1796). Here are h. For every square matrix 𝐴 and every scalar c, it is true that
two examples. det(c𝐴) = c det(𝐴).

1 a a2 a3 i. For all square matrices 𝐴 and 𝐵, it is true that

1 1 1 ⎡ ⎤
⎢1 b b2 b3 ⎥ det(𝐴 + 𝐵) = det(𝐴) + det(𝐵)
𝑉 = [a b c ] and 𝑉=⎢ 3⎥
⎢1 c c2 c ⎥
a2 b2 c2 j. For every 2 × 2 matrix 𝐴 it is true that
⎣1 d d2 d3 ⎦
det(𝐴2 ) = (det(𝐴))2
Vandermonde matrices arise in a variety of applications, such
as polynomial interpolation (see Formula (14) and Example 6
of Section 1.10). Use cofactor expansion to prove that Working with Technology
T1. a. Use the determinant capability of your technology utility
|1 x1 x 21 |
| | to find the determinant of the matrix
|1 x2 x 22 | = (x 2 − x 1 )(x 3 − x 1 )(x 3 − x 2 )
| | 4.2 −1.3 1.1 6.0
|1 x3 x 23 | ⎡ ⎤
⎢0.0 0.0 −3.2 3.4⎥
𝐴=⎢ ⎥
⎢4.5 1.3 0.0 14.8⎥
True-False Exercises ⎢ ⎥
⎣4.7 1.0 3.4 2.3⎦
TF. In parts (a)–( j) determine whether the statement is true or b. Compare the result obtained in part (a) to that obtained by
false, and justify your answer. a cofactor expansion along the second row of 𝐴.
a b
a. The determinant of the 2 × 2 matrix [ ] is ad + bc. T2. Let 𝐴n be the n × n matrix with 2’s along the main diagonal,
c d
1’s along the diagonal lines immediately above and below the
b. Two square matrices that have the same determinant main diagonal, and zeros everywhere else. Make a conjecture
must have the same size. about the relationship between n and det(𝐴n ).

2.2 Evaluating Determinants by

Row Reduction
In this section we will show how to evaluate a determinant by reducing the associated
matrix to row echelon form. In general, this method requires less computation than cofac-
tor expansion and hence is the method of choice for large matrices.

A Basic Theorem
We begin with a fundamental theorem that will lead us to an efficient procedure for eval-
uating the determinant of a square matrix of any size.

Theorem 2.2.1

Let 𝐴 be a square matrix. If 𝐴 has a row of zeros or a column of zeros, then

det(𝐴) = 0.
 2.2 Evaluating Determinants by Row Reduction 131

Solution By adding suitable multiples of the second row to the remaining rows, we obtain
|0 −1 1 3|
| |
1 2 −1 1|
det(𝐴) = ||
|0 0 3 3 ||
|0 1 8 0|

| −1 1 3|
| |
= −| 0 3 3| Cofactor expansion along
| | the first column
| 1 8 0|

| −1 1 3|
| |
= −| 0 3 3| We added the first row to
| | the third row.
| 0 9 3|
|3 3|
= −(−1) | | Cofactor expansion along
|9 3| the first column
= −18

Exercise Set 2.2

In Exercises 1–4, verify that det(𝐴) = det(𝐴𝑇 ). 1 3 1 5 3

⎡ ⎤
⎢−2 −7 0 −4 2⎥
−2 3 −6 1 ⎢ ⎥
1. 𝐴 = [ ] 2. 𝐴 = [ ] 13. ⎢ 0 0 1 0 1⎥
1 4 2 −2
⎢ 0 0 2 1 1⎥
⎢ ⎥
2 −1 3 4 2 −1 ⎣ 0 0 0 1 1⎦
3. 𝐴 = [1 2 4] 4. 𝐴 = [ 0 2 −3] 1 −2 3 1
⎡ ⎤
5 −3 6 −1 1 5 ⎢ 5 −9 6 3⎥
14. ⎢
⎢−1 2 −6 −2⎥⎥
In Exercises 5–8, find the determinant of the given elementary matrix
by inspection. ⎣ 2 8 6 1⎦
In Exercises 15–22, evaluate the determinant, given that
1 0 0 0
⎡ ⎤ 1 0 0 |a b c|
⎢0 1 0 0⎥ | |
5. ⎢ 6. [ 0 1 0] |d e 𝑓 | = −6
⎢0 0 −5 0⎥⎥
|
|g h i|
|
−5 0 1
⎣0 0 0 1⎦
|d e 𝑓| |g h i|
| | | |
15. | g h i| 16. | d e 𝑓|
1 0 0 0 1 0 0 0 | | | |
⎡ ⎤ ⎡ ⎤ |a b c| |a b c|
⎢0 0 1 0⎥ ⎢0 − 31 0 0⎥
7. ⎢ 8. ⎢
⎢0 1 0 0⎥⎥ ⎢0 0 1 0⎥
⎥ | 3a
|
3b 3c |
|
|a + d
|
b+e c+𝑓|
|
⎣0 0 0 1⎦ ⎣0 0 0 1⎦ 17. | −d −e −𝑓 | 18. | −d −e −𝑓 |
| | | |
| 4g 4h 4i | | g h i |
In Exercises 9–14, evaluate the determinant of the matrix
by first reducing the matrix to row echelon form and then using |a + g b+h c + i| | a b c |
| | | |
some combination of row operations and cofactor expansion. 19. | d e 𝑓 | 20. | 2d 2e 2𝑓 |
| | | |
| g h i | | g + 3a h + 3b i + 3c |
3 −6 9 3 6 −9
| −3a −3b −3c | | a b c |
9. [−2 7 −2] 10. [ 0 0 −2] | | | |
21. | d e 𝑓 | 22. | d e 𝑓|
0 1 5 −2 1 5 | | | |
| g − 4d h − 4e i − 4𝑓 | | 2a 2b 2c |

2 1 3 1 23. Use row reduction to show that

⎡ ⎤ 1 −3 0
⎢1 0 1 1⎥ |1 1 1 |
11. ⎢ 12. [−2 4 1]
⎢0 2 1 0⎥⎥
|
|a b
|
c | = (b − a)(c − a)(c − b)
5 −2 2 | 2 |
⎣0 1 2 3⎦ |a b2 c2 |
132 C H APT ER 2 Determinants

24. Verify the formulas in parts (a) and (b) and then make a con- 1 2 0 0 0
jecture about a general result of which these results are special ⎡ ⎤
⎢0 1 2 0 0⎥
cases. ⎢
32. 𝑀 = ⎢0 0 1 0 0⎥⎥
0 0 a13 ⎢ ⎥
a. det [0 a22 a23 ] = −a13 a22 a31 ⎢0 0 0 1 2⎥
⎣2 0 0 0 1⎦
a31 a32 a33
0 0 0 a14 33. Let 𝐴 be an n × n matrix, and let 𝐵 be the matrix that results
⎡ ⎤ when the rows of 𝐴 are written in reverse order. State a theo-
⎢0 0 a23 a24 ⎥
b. det ⎢ = a14 a23 a32 a41 rem that describes how det(𝐴) and det(𝐵) are related.
⎢0 a32 a33 a34 ⎥
⎥
⎣a41 a42 a43 a44 ⎦
34. Find the determinant of the following matrix.

In Exercises 25–28, confirm the identities without evaluating any of a b b b

the determinants directly. ⎡ ⎤
⎢b a b b⎥
| a1 b1 a1 + b1 + c1 | | a1 b1 c1 | ⎢b b a b⎥
| | | | ⎢ ⎥
25. | a2 b2 a2 + b2 + c2 | = | a2 b2 c2 |
| | | | ⎣b b b a⎦
| a3 b3 a3 + b3 + c3 | | a3 b3 c3 |

| a1 + b1 t
|
a 2 + b2 t a 3 + b3 t |
|
| a1
|
a2 a3 |
|
True-False Exercises
26. | a1 t + b1 a2 t + b 2 a3 t + b3 | = (1 − t 2 ) | b1 b2 b3 |
| | | | TF. In parts (a)–(f ) determine whether the statement is true or
| c1 c2 c3 | | c1 c2 c3 | false, and justify your answer.
| a1 + b1 a1 − b 1 c1 | | a1 b1 c1 | a. If 𝐴 is a 4 × 4 matrix and 𝐵 is obtained from 𝐴 by inter-
| | | | changing the first two rows and then interchanging the
27. | a2 + b2 a2 − b 2 c2 | = −2 | a2 b2 c2 |
| | | | last two rows, then det(𝐵) = det(𝐴).
| a3 + b3 a3 − b 3 c3 | | a3 b3 c3 |

| a1 b1 + ta1 c1 + rb1 + sa1 | | a1 a2 a3 | b. If 𝐴 is a 3 × 3 matrix and 𝐵 is obtained from 𝐴 by mul-

| | | |
28. | a2 b2 + ta2 c2 + rb2 + sa2 | = | b1 b2 b3 | tiplying the first column by 4 and multiplying the third
| | | |
| a3 b3 + ta3 c3 + rb3 + sa3 | | c1 c2 c3 | column by 34 , then det(𝐵) = 3 det(𝐴).

In Exercises 29–30, show that det(𝐴) = 0 without directly evaluat- c. If 𝐴 is a 3 × 3 matrix and 𝐵 is obtained from 𝐴 by adding
ing the determinant. 5 times the first row to each of the second and third rows,
−2 8 1 4 then det(𝐵) = 25 det(𝐴).
⎡ ⎤
⎢ 3 2 5 1⎥
29. 𝐴 = ⎢
⎢ 1 10 6 5⎥⎥
d. If 𝐴 is an n × n matrix and 𝐵 is obtained from 𝐴 by mul-
tiplying each row of 𝐴 by its row number, then
⎣ 4 −6 4 −3⎦
−4 1 1 1 1 n(n + 1)
⎡ ⎤ det(𝐵) = det(𝐴)
⎢ 1 −4 1 1 1⎥ 2
30. 𝐴 = ⎢ 1 1 −4 1 1⎥
⎢ ⎥ e. If 𝐴 is a square matrix with two identical columns, then
⎢ 1 1 1 −4 1⎥
det(𝐴) = 0.
⎣ 1 1 1 1 −4⎦

It can be proved that if a square matrix 𝑀 is partitioned into block f. If the sum of the second and fourth row vectors of a 6 × 6
triangular form as matrix 𝐴 is equal to the last row vector, then det(𝐴) = 0.
𝐴 0 𝐴 𝐶
𝑀=[ ] or 𝑀 = [ ] Working with Technology
𝐶 𝐵 0 𝐵
T1. Find the determinant of
in which 𝐴 and 𝐵 are square, then det(𝑀) = det(𝐴) det(𝐵). Use
this result to compute the determinants of the matrices in Exer- 4.2 −1.3 1.1 6.0
cises 31 and 32. ⎡ ⎤
⎢0.0 0.0 −3.2 3.4⎥
1 2 0 8 6 −9 𝐴=⎢ ⎥
⎡ ⎤ ⎢4.5 1.3 0.0 14.8⎥
⎢ 2 5 0 4 7 5⎥ ⎢ ⎥
⎢−1 ⎣4.7 1.0 3.4 2.3⎦
⎢ 3 2 6 9 −2⎥ ⎥
31. 𝑀 = ⎢ ⎥
⎢ 0 0 0 3 0 0⎥ by reducing the matrix to reduced row echelon form, and
⎢ 0 0 0 2 1 0⎥ compare the result obtained in this way to that obtained in
⎢ ⎥
⎣ 0 0 0 −3 8 −4⎦ Exercise T1 of Section 2.1.
142 C H APT ER 2 Determinants

OPTIONAL: We now have all of the machinery necessary to prove the following two
results, which we stated without proof in Theorem 1.7.1:

• Theorem 1.7.1(c) A triangular matrix is invertible if and only if its diagonal entries
are all nonzero.
• Theorem 1.7.1(d) The inverse of an invertible lower triangular matrix is lower tri-
angular, and the inverse of an invertible upper triangular matrix is upper triangular.

Proof of Theorem 1.7.1(c) Let 𝐴 = [aij ] be a triangular matrix, so that its diagonal entries
are
a11 , a22 , . . . , ann
From Theorem 2.1.2, the matrix 𝐴 is invertible if and only if
det(𝐴) = a11 a22 ⋅ ⋅ ⋅ ann
is nonzero, which is true if and only if the diagonal entries are all nonzero.

Proof of Theorem 1.7.1(d) We will prove the result for upper triangular matrices and
leave the lower triangular case for you. Assume that 𝐴 is upper triangular and invertible.
Since
1
𝐴−1 = adj(𝐴)
det(𝐴)
we can prove that 𝐴−1 is upper triangular by showing that adj(𝐴) is upper triangular or,
equivalently, that the matrix of cofactors is lower triangular. We can do this by showing
that every cofactor 𝐶 ij with i < j (i.e., above the main diagonal) is zero. Since
𝐶 ij = (−1)i+j 𝑀ij
it suffices to show that each minor 𝑀ij with i < j is zero. For this purpose, let 𝐵ij be the
matrix that results when the ith row and jth column of 𝐴 are deleted, so
𝑀ij = det(𝐵ij ) (10)
From the assumption that i < j, it follows that 𝐵ij is upper triangular (see Figure 1.7.1).
Since 𝐴 is upper triangular, its (i + 1)-st row begins with at least i zeros. But the ith row of
𝐵ij is the (i + 1)-st row of 𝐴 with the entry in the jth column removed. Since i < j, none of
the first i zeros is removed by deleting the jth column; thus the ith row of 𝐵ij starts with at
least i zeros, which implies that this row has a zero on the main diagonal. It now follows
from Theorem 2.1.2 that det(𝐵ij ) = 0 and from (10) that 𝑀ij = 0.

Exercise Set 2.3

In Exercises 1–4, verify that det(k𝐴) = kn det(𝐴). 2 1 0 1 −1 3

−1 2 2 2 5. 𝐴 = [3 4 0] and 𝐵 = [7 1 2]
1. 𝐴 = [ ]; k = 2 2. 𝐴 = [ ]; k = −4
3 4 5 −2 0 0 2 5 0 1

2 −1 3 −1 8 2 2 −1 −4
3. 𝐴 = [3 2 1]; k = −2 6. 𝐴 = [ 1 0 −1] and 𝐵 = [1 1 3]
1 4 5 −2 2 2 0 3 −1

1 1 1 In Exercises 7–14, use determinants to decide whether the given

matrix is invertible.
4. 𝐴 = [0 2 3]; k = 3
0 1 −2 2 5 5 2 0 3
In Exercises 5–6, verify that det(𝐴𝐵) = det(𝐵𝐴) and determine 7. 𝐴 = [−1 −1 0] 8. 𝐴 = [ 0 3 2]
whether the equality det(𝐴 + 𝐵) = det(𝐴) + det(𝐵) holds. 2 4 3 −2 0 −4
 2.3 Properties of Determinants; Cramer’s Rule 143

2 −3 5 −3 0 1 31. Use Cramer’s rule to solve for the unknown y without solving
9. 𝐴 = [0 1 −3] 10. 𝐴 = [ 5 0 6] for the unknowns x, z, and 𝑤.
0 0 2 8 0 3 4x + y + z + 𝑤 = 6
4 2 8 1 0 −1 3x + 7y − z + 𝑤 = 1
11. 𝐴 = [ −2 1 −4] 12. 𝐴 = [9 −1 4] 7x + 3y − 5z + 8𝑤 = −3
3 1 6 8 9 −1 x + y + z + 2𝑤 = 3

2 0 0 32. Let 𝐴x = b be the system in Exercise 31.

⎡ √2 − √7 0⎤
13. 𝐴 = [ 8 1 0] 14. 𝐴 = ⎢
⎢3√2 −3√7
⎥
0⎥
a. Solve by Cramer’s rule.
−5 3 6 ⎢ ⎥
⎣ 5 −9 0⎦ b. Solve by Gauss–Jordan elimination.
c. Which method involves fewer computations?
In Exercises 15–18, find the values of k for which the matrix 𝐴 is
invertible. 33. Let
a b c
k−3 −2 k 2 𝐴 = [d e 𝑓]
15. 𝐴 = [ ] 16. 𝐴 = [ ]
−2 k−2 2 k g h i
1 2 4 1 2 0 Assuming that det(𝐴) = −7, find
17. 𝐴 = [3 1 6] 18. 𝐴 = [k 1 k] a. det(3𝐴) b. det(𝐴−1 ) c. det(2𝐴−1 )
k 3 2 0 2 1
a g d
In Exercises 19–23, decide whether the matrix is invertible, and if so, d. det((2𝐴)−1 ) e. det [b h e]
use the adjoint method to find its inverse. c i 𝑓
2 5 5 2 0 3 34. In each part, find the determinant given that 𝐴 is a 4 × 4
19. 𝐴 = [−1 −1 0] 20. 𝐴 = [ 0 3 2] matrix for which det(𝐴) = −2.
2 4 3 −2 0 −4
a. det(−𝐴) b. det(𝐴−1 ) c. det(2𝐴𝑇 ) d. det(𝐴3 )
2 −3 5 2 0 0
21. 𝐴 = [0 1 −3] 22. 𝐴 = [ 8 1 0] 35. In each part, find the determinant given that 𝐴 is a 3 × 3
0 0 2 −5 3 6 matrix for which det(𝐴) = 7.

1 3 1 1 a. det(3𝐴) b. det(𝐴−1 )
⎡ ⎤
⎢2 5 2 2⎥ c. det(2𝐴−1 ) d. det((2𝐴)−1 )
23. 𝐴 = ⎢
1 3 8 9⎥
⎢ ⎥
⎣1 3 2 2⎦
Working with Proofs
In Exercises 24–29, solve by Cramer’s rule, where it applies. 36. Prove that a square matrix 𝐴 is invertible if and only if 𝐴𝑇 𝐴 is
24. 7x 1 − 2x 2 = 3 25. 4x + 5y =2 invertible.
3x 1 + x 2 = 5 11x + y + 2z = 3
x + 5y + 2z = 1 37. Prove that if 𝐴 is a square matrix, then
det(𝐴𝑇 𝐴) = det(𝐴𝐴𝑇 )
26. x − 4y + z = 6 27. x 1 − 3x 2 + x 3 = 4
4x − y + 2z = −1 2x 1 − x 2 = −2 38. Let 𝐴x = b be a system of n linear equations in n unknowns
2x + 2y − 3z = −20 4x 1 − 3x 3 = 0 with integer coefficients and integer constants. Prove that if
det(𝐴) = 1, the solution x has integer entries.
28. −x 1 − 4x 2 + 2x 3 + x4 = −32
2x 1 − x2 + 7x 3 + 9x 4 = 14 39. Prove that if det(𝐴) = 1 and all the entries in 𝐴 are integers,
−x 1 + x2 + 3x 3 + x4 = 11 then all the entries in 𝐴−1 are integers.
x1 − 2x 2 + x3 − 4x 4 = −4

29. 3x 1 − x 2 + x 3 = 4
True-False Exercises
−x 1 + 7x 2 − 2x 3 = 1 TF. In parts (a)–(l) determine whether the statement is true or
2x 1 + 6x 2 − x 3 = 5 false, and justify your answer.

30. Show that the matrix a. If 𝐴 is a 3 × 3 matrix, then det(2𝐴) = 2 det(𝐴).

cos 𝜃 sin 𝜃 0 b. If 𝐴 and 𝐵 are square matrices of the same size such that
𝐴 = [ − sin 𝜃 cos 𝜃 0] det(𝐴) = det(𝐵), then det(𝐴 + 𝐵) = 2 det(𝐴).
0 0 1 c. If 𝐴 and 𝐵 are square matrices of the same size and 𝐴 is
−1
invertible, then
is invertible for all values of 𝜃; then find 𝐴 using Theo-
rem 2.3.6. det(𝐴−1 𝐵𝐴) = det(𝐵)
144 C H APT ER 2 Determinants

d. A square matrix 𝐴 is invertible if and only if det(𝐴) = 0. in which 𝜖 > 0. Since det(𝐴) = 𝜖 ≠ 0, it follows from The-
𝑇 orem 2.3.8 that 𝐴 is invertible. Compute det(𝐴) for various
e. The matrix of cofactors of 𝐴 is precisely [adj(𝐴)] .
small nonzero values of 𝜖 until you find a value that produces
f. For every n × n matrix 𝐴, we have det(𝐴) = 0, thereby leading you to conclude erroneously that
𝐴 ⋅ adj(𝐴) = (det(𝐴))𝐼n 𝐴 is not invertible. Discuss the cause of this.

g. If 𝐴 is a square matrix and the linear system 𝐴x = 0 has T2. We know from Exercise 39 that if 𝐴 is a square matrix then
multiple solutions for x, then det(𝐴) = 0. det(𝐴𝑇 𝐴) = det(𝐴𝐴𝑇 ). By experimenting, make a conjec-
ture as to whether this is true if 𝐴 is not square.
h. If 𝐴 is an n × n matrix and there exists an n × 1 matrix b
such that the linear system 𝐴x = b has no solutions, then T3. The French mathematician Jacques Hadamard (1865–1963)
the reduced row echelon form of 𝐴 cannot be 𝐼n . proved that if 𝐴 is an n × n matrix each of whose entries sat-
isfies the condition |aij | ≤ 𝑀, then
i. If 𝐸 is an elementary matrix, then 𝐸x = 0 has only the
trivial solution.
|det(𝐴)| ≤ √nn 𝑀 n
j. If 𝐴 is an invertible matrix, then the linear system 𝐴x = 0
has only the trivial solution if and only if the linear system (Hadamard’s inequality). For the following matrix 𝐴, use
𝐴−1 x = 0 has only the trivial solution. this result to find an interval of possible values for det(𝐴),
and then use your technology utility to show that the value of
k. If 𝐴 is invertible, then adj(𝐴) must also be invertible. det(𝐴) falls within this interval.
l. If 𝐴 has a row of zeros, then so does adj(𝐴).
0.3 −2.4 −1.7 2.5
⎡ ⎤
Working with Technology ⎢0.2 −0.3 −1.2 1.4⎥
𝐴=⎢ ⎥
⎢2.5 2.3 0.0 1.8⎥
T1. Consider the matrix ⎢ ⎥
1 1 ⎣1.7 1.0 −2.1 2.3⎦
𝐴=[ ]
1 1+𝜖

Chapter 2 Supplementary Exercises

In Exercises 1–8, evaluate the determinant of the given matrix by (a) 12. Use the determinant to decide whether the matrices in Exer-
cofactor expansion and (b) using elementary row operations to intro- cises 5–8 are invertible.
duce zeros into the matrix.
In Exercises 13–15, find the given determinant by any method.
−4 2 7 −1
1. [ ] 2. [ ] |3 −4 a|
3 3 −2 −6 | 5 b − 3| | |
13. | | 14. | a2 1 2|
|b − 2 −3 | | |
−1 5 2 −1 −2 −3 |2 a−1 4|
3. [ 0 2 −1] 4. [ −4 −5 −6]
−3 1 1 −7 −8 −9 |0 0 0 0 −3 |
| |
|0 0 0 −4 0|
3 0 −1 −5 1 4
15. || 0 0 −1 0 0 ||
5. [1 1 1] 6. [ 3 0 2] |0 2 0 0 0|
0 4 2 1 −2 2 | |
|5 0 0 0 0|
3 6 0 1 −1 −2 −3 −4
⎡ ⎤ ⎡ ⎤
⎢−2 3 1 4⎥ ⎢ 4 3 2 1⎥ 16. Solve for x.
7. ⎢ 8. ⎢ ⎥
⎢ 1 0 −1 1⎥⎥ ⎢ 1 2 3 4⎥ |x −1 | |
|1 0 −3 |
|
⎣−4 −3 −2 −1⎦ | | = |2 x −6 |
⎣−9 2 −2 2⎦ |3 1 − x| | |
|1 3 x − 5|
9. Evaluate the determinants in Exercises 3–6 by using the arrow
technique (see Example 7 in Section 2.1).
In Exercises 17–24, use the adjoint method (Theorem 2.3.6) to find
10. a. Construct a 4 × 4 matrix whose determinant is easy to com- the inverse of the given matrix, if it exists.
pute using cofactor expansion but hard to evaluate using
elementary row operations. 17. The matrix in Exercise 1. 18. The matrix in Exercise 2.

b. Construct a 4 × 4 matrix whose determinant is easy to com-

19. The matrix in Exercise 3. 20. The matrix in Exercise 4.
pute using elementary row operations but hard to evaluate
using cofactor expansion.
21. The matrix in Exercise 5. 22. The matrix in Exercise 6.
11. Use the determinant to decide whether the matrices in Exer-
cises 1–4 are invertible. 23. The matrix in Exercise 7. 24. The matrix in Exercise 8.
 Chapter 2 Supplementary Exercises 145

25. Use Cramer’s rule to solve for x′ and y′ in terms of x and y. 34. a. In the accompanying figure, the area of the triangle 𝐴𝐵𝐶
x= 3 ′ 4 ′ can be expressed as
5x − 5y
4 ′ 3 ′ area 𝐴𝐵𝐶 = area 𝐴𝐷𝐸𝐶 + area 𝐶𝐸𝐹𝐵 − area 𝐴𝐷𝐹𝐵
y= 5x + 5y

26. Use Cramer’s rule to solve for x′ and y′ in terms of x and y. Use this and the fact that the area of a trapezoid equals 12
x = x′ cos 𝜃 − y′ sin 𝜃 the altitude times the sum of the parallel sides to show that
y = x′ sin 𝜃 + y′ cos 𝜃 | x1 y1 1|
1| |
27. By examining the determinant of the coefficient matrix, show area 𝐴𝐵𝐶 = | x2 y2 1|
2| |
that the following system has a nontrivial solution if and only | x3 y3 1|
if 𝛼 = 𝛽.
x + y + 𝛼z = 0 [Note: In the derivation of this formula, the vertices are
labeled such that the triangle is traced counterclockwise
x + y + 𝛽z = 0
proceeding from (x 1 , y1 ) to (x 2 , y2 ) to (x 3 , y3 ). For a clock-
𝛼x + 𝛽y + z = 0 wise orientation, the determinant above yields the negative
28. Let 𝐴 be a 3 × 3 matrix, each of whose entries is 1 or 0. What of the area.]
is the largest possible value for det(𝐴)? b. Use the result in (a) to find the area of the triangle with ver-
tices (3, 3), (4, 0), (−2, −1).
29. a. For the triangle in the accompanying figure, use trigonom-
etry to show that
b cos 𝛾 + c cos 𝛽 = a
C(x3, y3)
c cos 𝛼 + a cos 𝛾 = b
a cos 𝛽 + b cos 𝛼 = c B(x2, y2)
and then apply Cramer’s rule to show that A(x1, y1)
b2 + c2 − a2
cos 𝛼 =
2bc
b. Use Cramer’s rule to obtain similar formulas for cos 𝛽 and D E F
cos 𝛾.
FIGURE Ex-34

γ
b a
α β
c
35. Use the fact that

FIGURE Ex-29
21375, 38798, 34162, 40223, 79154

are all divisible by 19 to show that

|2 1 3 7 5|
30. Use determinants to show that for all real values of 𝜆, the only | |
solution of |3 8 7 9 8|
x − 2y = 𝜆x |3 4 1 6 2 ||
|
x − y = 𝜆y |4 0 2 2 3|
| |
is x = 0, y = 0. |7 9 1 5 4|
31. Prove: If 𝐴 is invertible, then adj(𝐴) is invertible and is divisible by 19 without directly evaluating the determinant.
1
[adj(𝐴)]−1 = 𝐴 = adj(𝐴−1 ) 36. Without directly evaluating the determinant, show that
det(𝐴)
32. Prove: If 𝐴 is an n × n matrix, then | sin 𝛼 cos 𝛼 sin(𝛼 + 𝛿) |
| |
det[adj(𝐴)] = [det(𝐴)] n−1 | sin 𝛽 cos 𝛽 sin(𝛽 + 𝛿) | = 0
| |
| sin 𝛾 cos 𝛾 sin(𝛾 + 𝛿) |
33. Prove: If the entries in each row of an n × n matrix 𝐴 add up
to zero, then the determinant of 𝐴 is zero. [Hint: Consider the 𝑇 a b
37. Let 𝑇 ∶ 𝑅2 → 𝑅 be the mapping (a, b, c, d) ⟶ det[ ]. Is
product 𝐴x, where x is the n × 1 matrix, each of whose entries c d
is one.] this a linear transformation? Justify your answer.