Solution Manual For Advanced Engineering Mathematics 7th Edition Peter Oneil

Solution manual for Advanced Engineering
Mathematics 7th edition, Peter Oâ€™Neil
To download the complete and accurate content document, go to:

https://testbankbell.com/download/solution-manual-for-advanced-engineering-mathe
matics-7th-edition-peter-oneil/
Solution manual for Advanced Engineering Mathematics 7th edition, Peter Oâ€™Neil
Chapter 9
Eigenvalues and
Diagonalization
9.1 Eigenvalues and Eigenvectors

1.
pA (λ) = |λI − A| = λ2 − 2λ − 5
√ √
with roots (eigenvalues of A) λ1 = 1+ 6 and λ2 = 1− 6. Corresponding
eigenvectors are
√ √
6 − 6
V1 = , V2 = .
2 2
The Gershgorin circles are of radius 3 about (1, 0) and radius 2 about
(1, 0). These enclose the eigenvalues.
2.
pA (λ) = λ2 − 2λ − 8,

0 6
λ1 = 4, V1 = , λ2 = −2, V2 = .
4 −1
The Gershgorin circle is of radius 1 about (4, 0). The other Gershgorin
”circle” has radius 0 and so is not really a circle. We may think of this as
a degenerate circle containing the eigenvalue −2 in its interior.
3.
pA (λ) = λ2 + 3λ − 10,

7 0
λ1 = −5, V1 = , λ2 = 2, V2 = .
−1 1
The Gershgorin circle has radius 1 and center (2, 0).
203
© 2012 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Visit TestBankBell.com to get complete for all chapters

204 CHAPTER 9. EIGENVALUES AND DIAGONALIZATION
4.
pA (λ) = λ2 − 10λ + 18,

√ 2√ √ 2√
λ1 = 5 + 7, V1 = , λ2 = 5 − 7, V2 = .
1− 7 1+ 7
The Gershgorin circles have radius 2, center (6, 0) and radius 3, center
(4, 0).
5. pA (λ) = λ2 − 3λ + 14,
√
√ −1 + 47i
λ1 = (3 + 14i)/2, V1 =
4
√
√ −1 − 14i
λ2 = (3 − 14i)/2, V2 = .
4
(2, 0).
6.
pA (λ) = λ2 ,
with roots λ1 = λ2 = 0. The only eigenvectors are nonzero scalar multiples
of
1
V1 = .
0
The Gershgorin circle has radius 1, center the origin.
7.
pA (λ) = λ3 − 5λ2 + 6λ,
⎛ ⎞ ⎛ ⎞ ⎛ ⎞
0 2 0
λ1 = 0, V1 = ⎝1⎠ , λ2 = 2, V2 = ⎝1⎠ , λ3 = 3, V3 = ⎝2⎠ .
0 0 3
The Gershgorin circle has radius 3, center the origin.
8.
pA (λ) = (λ + 1)(λ2 − λ − 7),
⎛ ⎞ ⎛ ⎞
0 √ 2√
λ1 = 1, V1 = ⎝0⎠ , λ2 = (1 + 29)/2, V2 = ⎝5 + 29⎠ ,
1 0
⎛ ⎞
√ 2√
λ3 = (1 − 29)/2, V3 = ⎝5 − 29⎠ .
0
The Gershgorin circles have radius 1, center (−2, 0), and radius 1, center
(3, 0).
9.1. EIGENVALUES AND EIGENVECTORS 205
9.
pA (λ) = λ2 (λ + 3),
⎛ ⎞ ⎛ ⎞
1 1
λ1 = −3, V1 = ⎝0⎠ , λ2 = λ3 = 0, V2 = ⎝0⎠ .
0 3
There is only one independent eigenvector associated with eigenvalue 0.
The Gershgorin circle has radius 2, center (−3, 0).
10.
pA (λ) = λ3 + 2λ,
⎛ ⎞ ⎛ ⎞ ⎛ ⎞
0 √ 1 √ 1
λ1 = 0, V1 = ⎝1⎠ , λ2 = 2i, V2 = ⎝ −1 √
⎠ , λ3 = − 2i, V3 = ⎝ −1 ⎠ .
√
0 −2 2i 2i
The Gershgorin circles have center (0, 0) and radii 1 and 2.
11.
pA (λ) = (λ + 14)(λ − 2)2 ,
⎛ ⎞
−16
λ1 = −14, V1 = ⎝ 0 ⎠
1
⎛ ⎞
0
λ2 = λ3 = 2, V2 = ⎝0⎠ ,
1
with only one independent eigenvector associated with the multiple eigen-
value λ2 . The Gershgorin circles have radius 1, center (−14, 0) and radius
3, center (2, 0).
12.
pA (λ) = (λ − 3)(λ2 + λ − 42),
⎛
⎞ ⎛ ⎞ ⎛ ⎞
0 30 0
λ1 = 6, V1 = ⎝ 1 ⎠ , λ2 = 3, V3 = ⎝−2⎠ , λ3 = −7, V3 = ⎝8⎠ .
−1 5 5
The Gershgorin circles have radius 9, center (−2, 0), and radius 5, center
(1, 0).
13.
pA (λ) = λ(λ2 − 8λ + 7),
⎛⎞ ⎛ ⎞ ⎛ ⎞
14 6 0
λ1 = 0, V1 ⎝ 7 ⎠ , λ2 = 1, V2 = ⎝0⎠ , λ3 = 7, V3 = ⎝0⎠
10 5 1
(7, 0).
14.
pA = λ2 (λ2 + 2λ − 1),
λ1 = λ2 = 0,
with two associated independent eigenvectors
⎛ ⎞ ⎛ ⎞
1 0
⎜2⎟ ⎜0⎟
V1 = ⎝ ⎠ , V 2 = ⎝ ⎟
⎜ ⎟ ⎜ .
0 1⎠
−1 0
For the other two eigenvalues,

⎛ ⎞ ⎛ ⎞
1√ 1√
√ ⎜1 + 2⎟ √ ⎜1 − 2⎟
λ3 = −1 + 2, V3 = ⎜ ⎟ ⎜
⎝ 0 ⎠ , λ4 = −1 − 2, V4 = ⎝ 0 ⎠ .
⎟
0 0
The Gershgorin circles have radius 1, center (−2, 0) and radius 2, center
(0, 0).
15.
pA (λ) = (λ − 1)(λ − 2)(λ2 + λ − 13),
⎛ ⎞ ⎛ ⎞
−2 0
⎜−11⎟ ⎜0⎟
λ1 = 1, V1 = ⎜ ⎟ ⎜ ⎟
⎝ 0 ⎠ , λ2 = 2, V2 = ⎝1⎠ ,
1 0
⎛√ ⎞ ⎛ √ ⎞
√ 53 − 7 √ − 53 − 7
−1 + 53 ⎜ ⎟ ⎜ ⎟
λ3 = , V3 = ⎜
0 ⎟ , λ4 = −1 − 53 , V4 = ⎜ 0 ⎟.
2 ⎝ 0 ⎠ 2 ⎝ 0 ⎠
2 2
The Gershgorin circles have radius 2, center (−4, 0) and radius 1 and
center (3, 0).
16.
pA (λ) = λ2 (λ − 1)(λ − 5),
⎛ ⎞ ⎛ ⎞
1 1
⎜−4⎟ ⎜0⎟
λ1 = 1, V1 = ⎜ ⎟ ⎜ ⎟
⎝ 0 ⎠ , λ2 = 5, V2 = ⎝0⎠ ,
0 0
⎛ ⎞
0
⎜0⎟
λ3 = λ4 = 0, V3 = ⎜ ⎟
⎝1⎠ .
0
9.1. EIGENVALUES AND EIGENVECTORS 207
There is only one independent eigenvector associated with the double

eigenvalue 0. The Gershgorin circles have radius 2, center (0, 0), radius 1,
center (2, 0), and radius 1, center (2, 0). One of the Gershgorin circles is
enclosed by the other in this example.
17.
pA (λ) = λ2 − 5λ,

1 −2
λ1 = 0, V1 = , λ2 = 5, V2 = .
2 1
By taking the dot product of the eigenvectors, we see that they are or-
thogonal.
18.
pA (λ) = λ2 − λ − 37,
√ √
so eigenvalues are λ1 = (1+ 149)/2 and λ2 = (1− 49)/2. Corresponding
eigenvectors are

10
√ 10
√
V1 = and V2 = .
7 + 149 7 − 149
These eigenvectors are orthogonal.
19.
pA (λ) = λ2 − 10λ − 23,
√ √
so eigenvalues are λ1 = 5 + 2 and λ2 = 5 − 2. Corresponding eigen-
vectors are √ √
1+ 2 1− 2
V1 = and V2 = .
1 1
20.
pA (λ) = λ2 + 9λ − 53
√ √
so the eigenvalues of A are λ1 = (−9 + 293)/2 and λ2 = (−9 − 293)/2.
Corresponding eigenvectors are

2√ 2√
V1 = and V2 = .
17 + 293 17 − 293
21.
pA (λ) = (λ − 3)(λ2 + 2λ − 1),
√ √
so eigenvalues are λ1 = 3, λ2 = 1 + 2 and λ3 = −1 − 2. Corresponding
eigenvectors are
⎛ ⎞ ⎛ √ ⎞ ⎛ √ ⎞
0 1+ 2 1− 2
V1 = ⎝0⎠ , V2 = ⎝ 1 ⎠ , and V3 = ⎝ 1 ⎠ .
1 0 0
These eigenvectors are mutually orthogonal.
√
(λ − 2)(λ2 − 2λ − 2), so eigenvalues are λ1 = 2, λ2 = 1 +
22. pA (λ) = √ 3,
λ3 = 1 − 3. Corresponding eigenvectors are
⎛ ⎞ ⎛ √ ⎞ ⎛ √ ⎞
0 −1 + 3 −1 − 3
V1 = ⎝−1⎠ , V2 = ⎝ 1 ⎠ and V3 = ⎝ 1 ⎠.
1 1 1
These eigenvectors are mutually orthogonal.
23. We know that AE = λE. Then
A2 E = A(AE) = A(λE) = λ(AE) = λ2 E.
This means that λ2 is an eigenvalue of E2 , with eigenvector E. The general

result follows now from an induction on k to show that Ak = λk E.
24. The characteristic polynomial of A is pA (λ) = |λI − A| = 0. The constant

term of this polynomial is obtained by setting λ = 0. This constant term
is equal to | − A|. This determinant is (−1)n |A|. Now λ = 0 is an
eigenvalue of A (root of the characteristic polynomial) exactly when the
constant term of pA (λ) is zero, and we now know that this occurs exactly
when |A| = 0.
9.2 Diagonalization
1.
pA (λ) = λ2 − 3λ + 4
√
is the√characteristic polynomial, with roots λ1 = (3 + 7i)/2 and λ2 =
(3 − 7i)/2. Corresponding eigenvectors are
√ √
−3 + 7i −3 − 7i
V1 = and V2 = .
8 8
The matrix √ √
−3 + 7i −3 − 7i
P=
8 8
diagonalizes A and
√
(3 + 7i)/2 0
√
P−1 AP = .
0 (3 − 7i)/2
If we wrote the eigenvectors in the other order in forming P, then the

columns of P−1 AP would be reversed.
2.
pA (λ) = λ2 − 8λ + 12,
9.2. DIAGONALIZATION 209
so the eigenvalues are λ1 = 2 and λ2 = 6. Corresponding eigenvectors are

−1 3
V1 = and V2 = .
1 1
We can diagonalize A with

−1 3
P= ,
1 1
obtaining
2 0
P−1 AP = .
0 6
3.
pA (λ) = λ2 − 2λ + 1,
so the eigenvalues are λ1 = λ2 = 1. Every eigenvector is a scalar multiple
of
0
1
so A is not diagonalizable.
4.
pA (λ) = λ2 − 4λ − 45,
so the eigenvalues are λ1 = −5 and λ2 = 9. Corresponding eigenvectors
are
1 3
V1 = and V2 = .
0 14
Then
1 3
P=
0 14
diagonalizes A and
−1 −5 0
P AP = .
0 9
5.
pA (λ) = λ(λ − 5)(λ + 2),
and the eigenvalues of A are λ1 = 0, λ2 = 5 and λ3 = −2. Corresponding
eigenvectors are
⎛ ⎞ ⎛ ⎞ ⎛ ⎞
0 5 0
V1 = ⎝1⎠ , V2 = ⎝1⎠ and V3 = ⎝−3⎠ .
0 0 2
Form ⎛ ⎞
0 5 0
P = ⎝1 1 −3⎠
0 0 2
Then ⎛ ⎞
0 0 0
P−1 AP = ⎝0 5 0 ⎠.
0 0 −2
6.
pA (λ) = λ(λ − 3λ − 2),
√ √
so the eigenvalues are λ1 = 0, λ2 = (3 + 17)/2 and λ3 = (3 − 17)/2.
Corresponding eigenvectors are
⎛ ⎞ ⎛ ⎞ ⎛ ⎞
−2 0 0
V1 = ⎝−3⎠ , V2 = ⎝ 4√ ⎠ and V3 = ⎝ 4√ ⎠ .
1 3 + 17 3 − 17
Let ⎛ ⎞
−2 0 0
P = ⎝−3 4√ 4√ ⎠ .
1 3 + 17 3 − 17
Then ⎛ ⎞
0 √0 0
P−1 AP = ⎝0 (3 + 17)/2 0
√
⎠.
0 0 (3 − 17)/2
7.
pA (λ) = (λ + 2)2 (λ − 1),
so eigenvalues and corresponding eigenvectors are
⎛ ⎞ ⎛ ⎞
0 −3
λ1 = 1, V1 = ⎝1⎠ , λ2 = λ3 = −2, V2 = ⎝ 1 ⎠ .
0 0
A does not have three linearly independent eigenvectors (the repeated

eigenvalue has only one independent eigenvector), and so is not diagonal-
izable.
8.
pA (λ) = (λ − 2)(λ2 − 4λ + 5),
so eigenvalues and eigenvectors are
⎛ ⎞ ⎛ ⎞ ⎛ ⎞
1 0 0
λ1 = 2, V1 = ⎝0⎠ , λ2 = 2 + i, V2 = ⎝1⎠ , λ3 = 2 − i, V3 = ⎝ 1 ⎠ .
0 i −i
Let ⎛ ⎞
1 0 0
P = ⎝0 1 1⎠
0 i −i
and then ⎛ ⎞
2 0 0
P−1 AP = ⎝0 2 + i 0 ⎠.
0 0 2−i
9.
pA (λ) = (λ − 1)(λ − 4)(λ2 + 5λ + 5),
√
so eigenvalues
√ and eigenvectors are λ1 = 1, λ2 = 4, λ3 = (−5 + 5)/2 and
λ4 = (−5 − 5)/2. Corresponding eigenvectors are
⎛ ⎞ ⎛ ⎞
1 0
⎜0⎟ ⎜1⎟
V1 = ⎜ ⎟ ⎜ ⎟
⎝0⎠ , V2 = ⎝0⎠ ,
0 0
⎛ ⎞ ⎛ ⎞
0
√ 0
√
⎜(2 − 3 5)/41⎟ ⎜(2 + 3 5)/41⎟
V3 = ⎜ √ ⎟ ⎜ √ ⎟
⎝ (−1 + 5)/2 ⎠ and V4 = ⎝ (−1 − 5)/2 ⎠ .
1 1
Let ⎛ ⎞
1 0 0
√ 0
√
⎜0 1 (2 − 3 √5)/41 (2 + 3 √5)/41⎟
P=⎜
⎝0
⎟.
0 (−1 + 5)/2 (−1 − 5)/2 ⎠
0 0 1 1
Then ⎛ ⎞
1 0 0 0
⎜0 4 0√ 0 ⎟
−1 ⎜
P AP = ⎝ ⎟.
0 0 (−5 + 5)/2 0√ ⎠
0 0 0 (−5 − 5)/2
10.
pA (λ) = (λ + 2)4 ,
so the eigenvalues of A are −2, with multiplicity 4. We find that there

are only three independent eigenvectors, namely
⎛ ⎞ ⎛ ⎞ ⎛ ⎞
0 0 0
⎜1⎟ ⎜0⎟ ⎜ ⎟
⎜ ⎟ , ⎜ ⎟ and ⎜0⎟ .
⎝0⎠ ⎝1⎠ ⎝0⎠
0 0 1
Since A does not have four linearly independent eigenvectors, A is not

diagonalizable.
11. Since P diagonalizes A, P−1 AP = D, a diagonal matrix having the eigen-

values of A along its main diagonal. Then A = PDP−1 , and
Ak = (PDP−1 )k
= (PDP−1 )(PDP−1 ) · · · (PDP−1 )
= PDk P−1 ,
with the interior pairings of P−1 P canceling.

12.
pA (λ = λ2 − λ − 18,
√ √
so the eigenvalues of A are (1 + 73)/2 and (1 − 73)/2. Form a matrix
P with corresponding eigenvectors are columns, yielding

−6
√ −6√
P= .
7 + 73 7 − 73
Compute √
1 7 − √73 6
P−1 = √ .
12 73 −7 − 73 −6
Then √
16 ((1 + 73)/2)16 √0
A =P P−1
0 ((1 − 73)/2)16

6(216 ) − 316 3(216 ) − 317
= .
−217 + 2(316 ) −216 + 6(316 )
13.
pA (λ) = λ2 + 6λ + 5,
so the eigenvalues are −1 and −5. Form P using corresponding eigenvec-
tors as columns:
4 0
P= .
1 1
Then
A18 = PAP−1

4 0 1 0 1/4 0
=
1 1 0 518 −1/4 1

1 0
= .
(1 − 518 )/4 518
√ √
14. A has eigenvalues −3 + 10 and −3 − 10. Form P using corresponding
eigenvectors as columns:

3√ 3√
P= .
1 − 10 1 + 10
Then √ √
(10 + √10)/60 −√ 10/20
P−1 = .
(10 − 10)/60 10/20
Then √ 31
(−3 + 10) 0√
A31 = P P−1
0 (−3 − 10)31

a b
= ,
c d
where
1 √ √ √ √
a= √ (1 + 10)(−3 + 10)31 + ( 10 − 1)(−3 − 10)31 ,
2 10
3 √ √ 31
b= √ (−3 + 10)31 + (3 + 10) ,
2 10
1 √ √ √ √
c= √ (−1 + 10)(−3 + 10)31 + ( 10 + 1)(3 + 10)31
2 10
3 √ √ 31
d= √ (−3 + 10)31 + (3 + 10) .
2 10
√ √
15. Eigenvalues of A are λ1 = 2 and λ2 = − 2, with corresponding eigen-
vectors √ √
2 − 2
V1 = , V2 = .
1 1
Let √ √
2 − 2
P= .
1 1
We find that √
P−1 = √2/4 1/2 .
− 2/4 1/2
Then
√ √ √ 43 √
43 2 − 2 ( 2) √0 √2/4 1/2
A =
1 1 0 (− 2)43 − 2/4 1/2

0 222
= .
221 0
16. Since A2 is diagonalizable, A2 has n linearly independent eigenvectors

X1 , · · · , Xn , with associated eigenvalues λ1 , · · · , λn , respectively. (These
eigenvalues need not be distinct). Now
pA2 (λ) = (A2 − λj In )Xj = O
for j = 1, 2, · · · , n. Then
pA2 (λ) = (A − λj In )(A + λj I n )

= pA ( λj )pA (− λj ) = O
for j = 1, 2, · · · , n. Then
pA ( λj ) = 0 or pA (− λj ) = 0.
But this means that λj or − λj is an eigenvalue of A with associated

eigenvector Xj . This implies that A has n linearly independent eigenvec-
tors and is therefore diagonalizable.
9.3 Some Special Matrices

In Problems 1 - 12, begin by finding an orthogonal set of eigenvectors. Since
any nonzero constant times an eigenvector is also an eigenvector, multiply each
eigenvector by the reciprocal of its magnitude to obtain an orthonormal set of
eigenvectors. The matrix Q is an orthogonal matrix that diagonalizes the given
matrix.
For Problems 1 - 6, orthogonal eigenvectors were requested in Problems 17
- 22 of Section 9.1, so for these problems all that is needed is to normalize these
eigenvectors.
1. In Problem 17 of Section 9.1 we found the orthogonal eigenvectors

1 −2
V1 = , V2 = .
2 1
√
Divide each by its length 5 and use the resulting orthonormal vectors as
columns of Q:
√ √
1 √5 −2/√ 5
Q= .
2/ 5 1/ 5
Q is an orthogonal matrix that diagonalizes A.
2. Previously we obtained the eigenvectors

10
√ 10
√
V1 = and V2 = .
7 + 149 7 − 149
Divide each eigenvector by its length to form

⎛ ⎞
√ 10 √ √ 10 √
298+14 149 298−14 149 ⎠
Q = ⎝ 7+√149 √ .
√ √ √ 7− 149 √
298+14 149 298−14 149
9.3. SOME SPECIAL MATRICES 215
3. We know the eigenvectors

√ √
1+ 2 1− 2
V1 = and V2 = .
1 1
Divide each by its length to form

⎛ √ √ ⎞
√1+ √2 √1− 2
√
Q = ⎝ 4+2 2 4−2 2 ⎠
√ 1√ √ 1 √
4+2 2 4−2 2
4. We have the eigenvectors

2√ 2√
V1 = and V2 = .
17 + 293 17 − 293
Normalize these eigenvectors to form

⎛ ⎞
√ 2
√ √ 2
√
⎝ 586−34 298 586+34
√ 298 ⎠ .
Q= √
√ 17− 298
√ √ 17+ 298
√
586−34 298 586+34 298
5. Eigenvectors are
⎛ ⎞ ⎛ √ ⎞ ⎛ √ ⎞
0 1+ 2 1− 2
V1 = ⎝0⎠ , V2 = ⎝ 1 ⎠ , and V3 = ⎝ 1 ⎠ .
1 0 0
Normalize these to form columns of Q:

⎛ √ √ ⎞
0 √1+ √ 2 √1− 2
√
⎜ 4+2 2 4−2 2 ⎟
Q=⎜ ⎝0 √4+2√2
1 √ 1 √ ⎟⎠.
4−2 2
1 0 0
6. Eigenvectors are
⎛ ⎞ ⎛ √ ⎞ ⎛ √ ⎞
0 −1 + 3 −1 − 3
V1 = ⎝−1⎠ , V2 = ⎝ 1 ⎠ and V3 = ⎝ 1 ⎠.
1 1 1
Normalize these to form

⎛ √ √ ⎞
−1+
√ 3 √ 3
−1−
0
⎜ √ √6 √ ⎟
6
Q = ⎝−1/ 2 1/√6 1/√6 ⎠ .
√
1/ 2 1/ 6 1/ 6
7.
pA (λ) = λ(λ2 − 5λ − 4)
√ √
and A has eigenvalues 0, (5 + 41)/2 and 5 − 41)/2, with corresponding
eigenvectors
⎛ ⎞ ⎛ √ ⎞ ⎛ √ ⎞
0 5 + 41 5 − 41
V1 = ⎝1⎠ , V2 = ⎝ 0 ⎠ , V3 = ⎝ 0 ⎠.
0 4 4
Normalize these to find an orthogonal matrix that diagonalizes A:

⎛ √ √ √ √ ⎞
0 (5 + 41)/ 82 + 10 41 (5 − 41)/ 82 − 10 41
⎜ ⎟
Q = ⎝1 0 0 ⎠.
√ √
0 4/ 82 + 10 41 4/ 82 − 10 41
8.
pA (λ) = λ(λ2 − 2λ − 16)
√ √
so 0, 1 + 17 and 1 − 17 are eigenvalues. We find the corresponding
eigenvectors
⎛ ⎞ ⎛ √ ⎞ ⎛ √ ⎞
0 1 + 17 1 − 17
V1 = ⎝0⎠ , V2 = ⎝ −4 ⎠ , V3 = ⎝ −4 ⎠ .
1 0 0
Then
⎛ √ √ √ √ ⎞
0 (1 + 17)/ 34 + 2 17 (1 − 17)/ 34 − 2 17
⎜ √ √ ⎟
Q = ⎝0 −4/ 34 + 2 17 −4/ 34 − 2 17 ⎠.
1 0 0
9.
pA (λ) = λ(λ2 − λ − 4)
√ √
and the eigenvalues are 0, (1 + 17)/2, (1 − 17)/2. Corresponding
eigenvectors are
⎛ ⎞ ⎛ ⎞ ⎛ ⎞
1 0√ 0√
⎝0⎠ , V2 ⎝−1 − 17⎠ , V3 = ⎝−1 − 17⎠ .
0 4 4
Then
⎛ ⎞
1 0 0
⎜ √ √ √ √ ⎟
Q = ⎝0 (−1 − 17)/ 34 + 2 17 (−1 + 17)/ 34 + 2 17⎠ .
√ √
0 4/ 34 + 2 17 4/ 34 + 2 17
10.
pA (λ) = (λ − 1)(λ2 − λ − 10)
√ √
and the eigenvalues are 1, (1 + 41)/2 and (1 − 41)/2. Corresponding
eigenvectors are
⎛ ⎞ ⎛ ⎞ ⎛ ⎞
1 6√ 6√
V1 = ⎝ 0 ⎠ , V2 = ⎝−1 + 41⎠ , V3 = ⎝−1 − 41⎠ .
−3 2 2
Then
⎛ √ √ √ ⎞
1/ 10 6/ 82 − 2 41 6/ 82 + 2 41
⎜ √ √ √ √ ⎟
Q=⎝ 0 (−1 + 41)/ 82 − 2 41 (−1 − 41)/ 82 − 2 41⎠ .
√ √ √
−2/ 10 2/ 82 − 2 41 2/ 82 − 2 41
11.
pA (λ) = λ2 (λ2 − 2λ − 3)
so the eigenvalues are 0, 0, −1 and 3. Corresponding eigenvectors are
⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞
1 0 0 0
⎜ 0⎟ ⎜0⎟ ⎜1⎟ ⎜−1⎟
V 1 = ⎝ ⎠ , V 2 = ⎝ ⎠ , V3 = ⎝ ⎠ , V 4 = ⎝ ⎟
⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ .
0 0 1 1⎠
0 1 0 0
Then ⎛ ⎞
1 0 0√ 0√
⎜0 0 1/ 2 −1/√ 2⎟
Q=⎜ √
⎝0 0 1/ 2
⎟.
1/ 2 ⎠
0 1 0 0
12.
pA = λ(λ − 5)(λ2 − 1)
and the eigenvalues are 0, 5, 1 and −1. Corresponding eigenvectors are
⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞
0 1 0 0
⎜ 0⎟ ⎜0⎟ ⎜1⎟ ⎜1⎟
V 1 = ⎝ ⎠ , V 2 = ⎝ ⎠ , V4 = ⎝ ⎠ , V 4 = ⎝ ⎟
⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ .
0 0 −1 1⎠
1 0 0 0
Then ⎛ ⎞
0 1 0√ 0√
⎜0 0 1/ 2 1/ 2⎟
Q=⎜ √ √ ⎟
⎝0 0 −1/ 2 1/ 2⎠ .
1 0 0 0
13. The matrix is not hermitian, skew-hermitian or unitary. Compute
pA (λ) = (λ − 2)2 .
The eigenvalue is 2 with multiplicity 2 and only one independent eigen-

vector,
i
.
1
Therefore A is not diagonalizable.
14. A is not hermitian, skew-hermitian or unitary.
pA (λ) = (λ + 1)2
with root −1 of multiplicity 2. Every eigenvector is a scalar multiple of

i
,
−1
so the matrix is not diagonalizable.

15. This matrix S is skew-hermitian, since
St = −S.
ps (λ) = λ(λ2 + 3)
√ √
so the eigenvalues are 0, 3i and − 3i, with corresponding eigenvectors
⎛ ⎞⎛ ⎞ ⎛ ⎞
2 √1 √1
⎝ 0 ⎠ ⎝ 3i ⎠ , and ⎝ − 3i ⎠ .
1 + i, −1 − i −1 − i
Let P have these eigenvectors as columns (in the given order). Then
⎛ ⎞
0 √0 0
P−1 SP = ⎝0 3i √0 ⎠.
0 0 − 3i
16. It is routine to check that

UUt = I,
so U is unitary.

1+i 2
pU (λ) = (λ − 1) λ − √ λ + i
2
so the eigenvalues are
√ √ √ √
1+ 3 1− 3 1− 3 1+ 3
1, √ + √ i, √ + √ i,
2 2 2 2 2 2 2 2
with corresponding eigenvectors

⎛ ⎞ ⎛ ⎞ ⎛ ⎞
0 1 +√i 1 +√i
⎝0⎠ , ⎝(1 − 3)i⎠ and ⎝(1 + 3)i⎠ .
1 0 0
Use these as the columns of P. Then P diagonalizes U.

17. The matrix is hermitian, since Ht = H. The eigenvalues are approxi-
mately λ1 = 4.051374, λ2 = 0.482696, λ3 = −1.53407. These are distinct,
so the matrix is diagonalizable.
√ √
18. H is hermitian with eigenvalues 1, (−1 + 41)/2 and (−1 − 41)/2. Cor-
responding eigenvectors are
⎛ ⎞ ⎛ ⎞ ⎛ ⎞
0 6 − 2i 6 − 2i
⎝1⎠ , ⎝ 0√ ⎠ and ⎝ 0√ ⎠ .
0 1 + 41 1 − 41
The matrix having these eigenvectors as columns diagonalizes H.

19. The matrix S is skew-hermitian with approximate eigenvalues −2.164248i,
0.772866i and 2.39182i. Since these are distinct, S is diagonalizable.
20. The matrix is not hermitian, skew-hermitian or unitary. Eigenvalues are
1, −1 and 3i, with corresponding eigenvectors
⎛ ⎞ ⎛ ⎞ ⎛ ⎞
0 0 −10i
⎝ i ⎠ , ⎝ i ⎠ and ⎝ 3 ⎠ .
1 −1 −1
The matrix having these eigenvectors as columns diagonalizes the matrix.

√ √
21. H is hermitian with eigenvalues 0, 4 + 3 2 and 4 − 3 2. Corresponding
eigenvectors are
⎛ ⎞ ⎛ √ ⎞ ⎛ √ ⎞
0 4+3 2 4−3 2
⎝ i ⎠ , ⎝ −1 ⎠ and ⎝ −1 ⎠ .
1 −i −i
The matrix having these eigenvectors as columns diagonalizes H.

22. The matrix of the quadratic form is

−5 2
A= .
2 3
√ √
Eigenvalues of A are −1 + 2 5 and −1 − 2 5 and the standard form of
this quadratic form is
√ √
(−1 + 2 5)y12 + (−1 − 2 5)y22 .
23. The matrix of this form is

4 −6
A= .
−6 1
√ √
The eigenvalues are (5 + 153)/2 and (5 − 153)/2. The standard form
is √ √
5 + 153 2 5 − 153
y1 + y22 .
2 2
24. The matrix is

−3 2
2 7
√
with eigenvalues 2 ± 29. The standard form is
√ √
(2 + 29)y12 + (2 − 29)y22 .
25. The matrix is

−2 4
1 −2
√ √
with eigenvalues (3 + 17)/2 and (3 − 17)/2. The standard form is
√ √
3 + 17 2 3 − 17
y1 + y22 .
2 2
26. The matrix is

0 −3
−3 4
√
√ √
(2 + 13)y12 + (2 − 13)2 y22 .
27. The matrix is

5 2
2 2
with eigenvalues 1, 6. The standard form is
y12 + 6y22 .
28. The matrix is

0 −1
−1 2
√
√ √
(1 + 2)y12 + (1 − 2)y22 .
In computations involving complex matrices, it is assumed that the conjugate

of a product is the product of the conjugates, and the conjugate of a transpose
is the transpose of the conjugate.
29. If A is hermitian, then At = A, so
(AAt ) = A(A)t = A(At )t = AA.
30. If H is hermitian, then Ht = H, so the diagonal elements satisfy hjj = hjj ,

and therefore must be real.
31. If S is skew-hermitian, then St = −S, so sjj = −sjj for j = 1, · · · , n. Now
write sjj = ajj + ibjj . Then ajj = −ajj , so each ajj = 0. This makes the
diagonal elements zero (if b = 0) or pure imaginary (if b = 0).
t t
32. Let U and V be unitary matrices. Then U−1 = U and V−1 = V . Then
t t
(UV)−1 = V−1 U−1 = V U = ((U)(V))t = (UV)t
and this implies that UV is unitary.
Solution manual for Advanced Engineering Mathematics 7th edition, Peter Oâ€™Neil
Visit TestBankBell.com to get complete for all chapters

Solution Manual For Advanced Engineering Mathematics 7th Edition Peter Oneil

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Solution Manual For Advanced Engineering Mathematics 7th Edition Peter Oneil

Uploaded by

Copyright:

Available Formats

Solution manual for Advanced Engineering

Mathematics 7th edition, Peter Oâ€™Neil

To download the complete and accurate content document, go to:

9.1 Eigenvalues and Eigenvectors

Visit TestBankBell.com to get complete for all chapters

For the other two eigenvalues,

There is only one independent eigenvector associated with the double

These eigenvectors are mutually orthogonal.

23. We know that AE = λE. Then

A2 E = A(AE) = A(λE) = λ(AE) = λ2 E.

This means that λ2 is an eigenvalue of E2 , with eigenvector E. The general

24. The characteristic polynomial of A is pA (λ) = |λI − A| = 0. The constant

If we wrote the eigenvectors in the other order in forming P, then the

so the eigenvalues are λ1 = 2 and λ2 = 6. Corresponding eigenvectors are

A does not have three linearly independent eigenvectors (the repeated

so the eigenvalues of A are −2, with multiplicity 4. We ﬁnd that there

Since A does not have four linearly independent eigenvectors, A is not

11. Since P diagonalizes A, P−1 AP = D, a diagonal matrix having the eigen-

with the interior pairings of P−1 P canceling.

16. Since A2 is diagonalizable, A2 has n linearly independent eigenvectors

pA2 (λ) = (A2 − λj In )Xj = O

pA2 (λ) = (A − λj In )(A + λj I n )

But this means that λj or − λj is an eigenvalue of A with associated

9.3 Some Special Matrices

1. In Problem 17 of Section 9.1 we found the orthogonal eigenvectors

2. Previously we obtained the eigenvectors

Divide each eigenvector by its length to form

3. We know the eigenvectors

Divide each by its length to form

4. We have the eigenvectors

Normalize these eigenvectors to form

Normalize these to form columns of Q:

Normalize these to form

Normalize these to ﬁnd an orthogonal matrix that diagonalizes A:

13. The matrix is not hermitian, skew-hermitian or unitary. Compute

The eigenvalue is 2 with multiplicity 2 and only one independent eigen-

with root −1 of multiplicity 2. Every eigenvector is a scalar multiple of

so the matrix is not diagonalizable.

16. It is routine to check that

with corresponding eigenvectors

Use these as the columns of P. Then P diagonalizes U.

The matrix having these eigenvectors as columns diagonalizes H.

The matrix having these eigenvectors as columns diagonalizes the matrix.

The matrix having these eigenvectors as columns diagonalizes H.

23. The matrix of this form is

24. The matrix is

25. The matrix is

26. The matrix is

27. The matrix is

28. The matrix is

In computations involving complex matrices, it is assumed that the conjugate

29. If A is hermitian, then At = A, so

(AAt ) = A(A)t = A(At )t = AA.

30. If H is hermitian, then Ht = H, so the diagonal elements satisfy hjj = hjj ,

and this implies that UV is unitary.

222 CHAPTER 9. EIGENVALUES AND DIAGONALIZATION

Visit TestBankBell.com to get complete for all chapters

You might also like