Linear Algebra: Homework 7

Alvin Lin
August 2016 - December 2016

Section 3.5
Exercise 1
 
x
Let S be the collection of vectors in R2 that satisfy the given property. In each case, either prove that
y
S forms a subspace of R2 or give a counterexample to show that it does not.
x=0
       
x 0 0 0
S= = =y = span
y y 1 1
 
0
Since ∈ R2 , S is a subspace of R2 .
1

Exercise 3
 
x
Let S be the collection of vectors in R2 that satisfy the given property. In each case, either prove that
y
S forms a subspace of R2 or give a counterexample to show that it does not.
y = 2x
       
x x 1 1
S= = =x = span
y 2x 2 2
 
1
Since ∈ R2 , S is a subspace of R2 .
2

Exercise 5
 
x
Let S be the collection of vectors y  in R3 that satisfy the given property. In each case, either prove
z
that S forms a subspace of R2 or give a counterexample to show that it does not.
x=y=z
       
x x 1 1
S = y = x = x 1 = span
       1
z x 1 1
 
1
Since 1 ∈ R3 , S is a subspace of R3 .
1

1
Exercise 7
 
x
Let S be the collection of vectors y  in R3 that satisfy the given property. In each case, either prove

z
2
that S forms a subspace of R or give a counterexample to show that it does not.
x−y+z =1
0 − 0 + 0 6= 1
Since the zero vector is not in S, S is not a subspace of R3 .

Exercise 9
Prove that every line through the origin in R3 is a subspace of R3 .
     
x a1 a1
y  = t a2  = span a2 
z a3 a3
 
a1
Every line through the origin can be described as the span of a vector a2  ∈ R3 . Therefore, every line

a3
3
through the origin is a subspace of R .

Exercise 10
Suppose S consists of all points in R2 that are on the x-axis or the y-axis (or both). Is S a subspace of
R2 ? Why or why not?
     
x 0
S= |x∈R ∪ |y∈R
0 y
 
1
∈S
0
 
0
∈S
1
     
1 0 1
+ = ∈
/S
0 1 1
The subspace S is not closed under addition.

Exercise 15
 
−1
If A is the matrix in Exercise 11, is ~v = 3  in null(A)?

−1
 
1 0 −1
A=
1 1 1
 
1 0 −1  
A~v = −1 3 −1
1 1 1
 
0
=
1
∴ ~v ∈
/ null(A)

2
Exercise 16
 
7
If A is the matrix in Exercise 12, is ~v = −1 in null(A)?
2
 
1 1 −3
A = 0 2 1
1 −1 −4
  
1 1 −3 7
A~v = 0 2
 1  −1
1 −1 −4 2
 
0
= 0

0
∴ ~v ∈ null(A)

Exercise 17
Give bases for row(A), col(A), and null(A).
 
1 0 −1
A=
1 1 1
 
1 0 −1
=
0 1 2
   
basis for row(A) = 1 0 −1 , 0 1 2
   
1 0
basis for col(A) = ,
1 1
 
x1
A x2  = ~0

x3
1x1 − 1x3 = 0
1x2 + 2x3 = 0
 
x1
null(A) = −2x1 
x1
 
1
= x1 −2
1
  
 1 
basis for null(A) = −2
1
 

3
Exercise 19
Give bases for row(A), col(A), and null(A).
 
1 1 0 1
A = 0 1 −1 1 
0 1 −1 −1
 
1 1 0 1
= 0 1 −1 1 
0 0 0 −2
 
1 1 0 0
= 0 1 −1 0
0 0 0 1
 
1 0 1 0
= 0 1 −1 0
0 0 0 1
     
basis for row(A) = 1 0 1 0 , 0 1 −1 0 , 0 0 0 1
     
 1 1 1 
basis for col(A) = 0 , 1 ,  1 
0 1 −1
 

A~x = 0
x1 + x3 = 0
x2 − x3 = 0
x4 = 0
 
x1
−x1 
null(A) = 
−x1 


0
  

 1 
  
−1

basis for null(A) =  


 −1 
0
 

Exercise 27
Find a basis for the span of the given vectors.
     
1 −1 0
−1 0 1
0 1 −1

4
    
1 −1 0 1 −1 0
−1 0 1  = 0 −1 1 
0 1 −1 0 1 −1
 
1 −1 0
= 0 1 −1
0 0 0
   
 1 −1 
basis = −1 ,  0 
0 1
 

Exercise 29
Find a basis for the span of the given vectors.
     
2 −3 −1 1 −1 0 4 −4 1

   
2 −3 −1 0 −1 −1
1 −1 0  = 1 −1 0 
4 −4 1 0 0 1
 
1 0 0
= 0
 1 0
0 0 1
basis = {e1 , e2 , e3 }

Exercise 35
Give the rank and nullity of the matrix in Exercise 17.
 
1 0 −1
A=
1 1 1
 
1 0 −1
=
0 1 2
rank(A) = 2
nullity(A) = 3 − rank(A) = 1

Exercise 37
Give the rank and nullity of the matrix in Exercise 19.
 
1 1 0 1
A= 0  1 −1 1 
0 1 −1 −1
 
1 0 1 0
= 0 1 −1 0
0 0 0 1
rank(A) = 3
nullity(A) = 4 − rank(A) = 1

5
Exercise 39
If A is a 3 × 5 matrix, explain why the columns of A must be linearly dependent.

rank(A) ≤ min(m, n)
A~x = ~0
5 = rank(A) + nullity(A)
rank(A) ≤ 3

The nullity of A must be at least 2, therefore A~x = ~0 has a non-trivial solution. Thus, the columns of A
are linearly dependent. Since there are 5 columns and only 3 can be linearly independent, they must be
linearly dependent.

Exercise 40
If A is a 4 × 2 matrix, explain why the rows of A must be linearly dependent.
There are four rows in total, but only two rows can be linearly independent in a 4 × 2 matrix, therefore
the rows must be linearly dependent.

Exercise 41
If A is a 3 × 5 matrix, what are the possible values of nullity(A)?

rank(A) ≤ 3
n = rank(A) + nullity(A)
5 = rank(A) + nullity(A)
5 − nullity(A) ≤ 3
−nullity(A) ≤ −2
nullity(A) ≥ 2
2 ≤ nullity(A) ≤ 5

Exercise 42
If A is a 4 × 2 matrix, what are the possible values of nullity(A)?

rank(A) ≤ 2
n = rank(A) + nullity(A)
2 = rank(A) + nullity(A)
2 − nullity(A) ≤ 2
−nullity(A) ≤ 0
nullity(A) ≥ 0
0 ≤ nullity(A) ≤ 2

6
Exercise 43
Find all possible values of rank(A) as a varies.
 
1 2 a
A = −2 4a 2
a −2 1
 
1 2 a
= 0 4a + 4 2 + 2a
1 − a2 1
a
 
1 2 a
= 0 4a + 4 2 + 2a
0 − a2 − 2 a1 − a
 
1 2 a
= 0 4a + 4 2 + 2a
0 −2 − 2a 1 − a2
 
1 2 a
= 0 2a + 2 1 + a 
0 2a + 2 a2 − 1
 
1 2 a
= 0 2a + 2 1+a 
2
0 0 a −a−2
2a + 2 = 0
a = −1
 
1 2 −1
→ 0 0 0 
0 0 2
a2 − a − 2 = 0
(a − 2)(a + 1) = 0
a = 1, −2
 
1 2 1
→ 0 4 2
0 0 0
 
1 2 a
→ 0 −2 −1
0 0 0

1 if a = 1

rank(A) = 2 if a = −1 ∨ a = 2

3 otherwise


7
Exercise 44
Find all possible values of rank(A) as a varies.
 
a 2 −1
A= 3 3 −2
−2 −1 a
2
− a1
 
1 a
= 1 2 −2 + a
−2 −1 a
 2
− a1

1 a
= 1 2 −2 + a 
0 3 a + 2(−2 + a)
2
− a1
 
1 a
= 0 2 − a2 −2 + a + a1 
0 3 3a − 4
 
a 2 −1
= 0 2a − 2 −2a + a2 + 1
0 3 3a − 4
a=1
 
1 2 −1
→ 0
 0 0
0 3 −1
(
2 if a = 1
rank(A) =
3 otherwise

Exercise 45
     
1 1 0
Do 1 , 0 , 1 form a basis for R3 ?
0 1 1
   
1 1 0 1 0 −1
1 0 1 = 1 0 1
0 1 1 0 1 1
 
1 0 −1
= 0 0 2
0 1 1
 
1 0 0
= 0 0 1
0 1 0
 
1 0 0
= 0
 1 0
0 0 1

Since the vectors are linearly independent and form the standard basis vectors, they form a basis for R3 .

8
Exercise 47
       
1 1 1 0
1 1 0 1
Do 
1 , 0 , 1 , 1 for a basis for
       R4 ?
0 1 1 1
   
1 1 1 0 1 1 1 0
1
 1 0 1
=
0 0 −1 1 
1 0 1 1 0 −1 0 1
0 1 1 1 0 1 1 1
 
1 1 1 0
0 0 −1 1
=
0

0 1 2
0 1 1 1
 
1 1 1 0
0 0 0 3
=
0

0 1 2
0 1 1 1
 
1 0 0 0
0 0 0 1
=
0

0 1 0
0 1 0 0
 
1 0 0 0
0 1 0 0
=
0

0 1 0
0 0 0 1

Since the vectors form the standard basis vectors, they form a basis for R4 .

Exercise 51
Show that w is in span(B) and find the coordinate vector [w]B .
     
 1 1  1
B = 2 ,  0  w~ = 6
0 −1 2
 

     
1 1 1
c1 2 + c2  0  = 6
0 −1 2
   
1 1 1 1 1 1
2 0 6 = 0 −2 4
0 −1 2 0 −1 2
 
1 0 3
= 0 1 −2
0 0 0
 
3
[w]B =
−2

9
Exercise 52
Show that w is in span(B) and find the coordinate vector [w]B .
     
 3 5  1
B = 1 , 1 ~ = 3
w
4 6 4
 

     
3 5 1
c1 1 + c2 1 = 3
4 6 4
   
3 5 1 0 2 −8
1 1 3 = 1 1 3
4 6 4 2 3 2
 
0 1 −4
= 1 1
 3
0 1 −4
 
1 0 7
= 0 1
 −4
0 0 0
 
7
[w]B =
−4

Exercise 57
If A is m × n, prove that every vector in null(A) is orthogonal to every vector in row(A).

A~x = ~0 ∀~x ∈ Rm
Xm
~y = ci ai ∀~u ∈ row(A)
i=1
 
c1
= col1 (A) . . . colm (A)  ... 
  

cm
 
c1
T  .. 
=A  . 
cm
xT y = xT AT c
xT y = (Ax)T c
x · y = 0c = 0

10
Exercise 58
If A and B are n × n of rank n, prove that AB has rank n.

nullity(A) = 0
nullity(B) = 0
B~x = 0
A(B~x) = 0
(AB)~x = 0
nullity(AB) = 0
rank(AB) = n − nullity(AB) = n

Exercise 59a
Prove that rank(AB) ≤ rank(B).

n = rank(AB) + nullity(AB)
= rank(B) + nullity(B)
nullity(B) = dim(null(B))
≤ dim(null(AB))
≤ nullity(AB)
rank(AB) + nullity(AB) = rank(B) + nullity(B)
∴ rank(AB) ≤ rank(B)

Exercise 60a
Prove that rank(AB) ≤ rank(A).
 
AB = A col1 (B) . . . coln (B)
col(AB) ⊆ col(B)
dim(col(AB)) ≤ dim(col(B))
rank(AB) ≤ rank(B)

Exercise 61
Prove that if U is invertible, then rank(U A) = rank(A).

A = IA
A = U −1 U A
rank(U ) = n
∴ rank(A) = n = rank(U )

Prove that if V is invertible, then rank(AV ) = rank(A).

A = AI
A = AV −1 V
rank(U ) = n
∴ rank(A) = n = rank(V )

11
Exercise 62
Prove that an m × n matrix A has rank 1 if and only if A can be written as the outer product uv T of a
vector u ∈ Rm and v ∈ Rn .
Suppose:
   
a1 c1
 ..   .. 
A =  .  =  . w
an cn
 
c1
 .. 
u=.
cn
w = vT
A = uv T
rank(A) = rank(uv T )
≤ rank(u)
≤1
rank(A) = 1

Exercise 63
Prove that an m × n matrix A has rank r, prove that A can be written as the sum of r matrices, each of
which has rank 1.

rank(A) = r
= dim(row(A))
 
a1
 .. 
A= . 
am
 
c11 v1 + c12 v2 + · · · + c1r vr
 c21 v1 + c22 v2 + · · · + c2r vr 
=
 
.. 
 . 
cm1 v1 + vm2 v2 + · · · + cmr vr
   
c11 c1r
 c21   c2r 
=  ..  v1 + · · · +  ..  vr
   
 .   . 
cm1 cmr

 
c11
 c21 
Because  .. , has rank 1, each of the r matrices has rank 1.
 
 . 
cm1

12
Exercise 64
Prove that, for m × n matrices A and B, rank(A + B) ≤ rank(A) + rank(B).

rowi (A + B) = rowi (A) + rowi (B)

The rows of A + B can be expressed as linear combinations of the respective rows of A and B.

Exercise 65
Let A be an n × n matrix such that A2 = 0. Prove that rank(A) ≤ n2 .

A2 = A col1 (A) . . . coln (A)

 

=0
A~x = 0
col(A) ⊆ null(A)
rank(A) + nullity(A) = n
rank(A) + rank(A) ≤ n
2rank(A) ≤ n
n
rank(A) ≤
2

Exercise 66
Let A be a skew-symmetric n × n matrix.

• Prove that xT Ax = 0 for all x ∈ Rn .

xT Ax = (xT Ax)T )
= (Ax)T (xT )T
= xT A T x
= xT (−A)x
xT Ax = −xT Ax
∴ xT Ax = 0

• Prove that I + A is invertible.

If this is true (I + A)x = 0 has only the trivial solution.

(I + A)x = 0
x + Ax = 0
xT x + xT Ax = 0(xT )
xT x + 0 = 0
∴x=0

13
Section 3.6
Exercise 1
 
2 −1
Let TA : R2 → R2 be the matrix transformation corresponding to A = . Find TA (~u) and TA (~v ),
    3 4
1 3
where ~u = and ~v = .
2 −2
 
0
TA (~u) =
11
 
8
TA (~v ) =
1

Exercise 3
Prove that the given transformation is a linear transformation, using the definition.
   
x x+y
T =
y x−y
 
ux + vx
T (u + v) = T
uy + vy
 
ux + vx + uy + vy
=
ux + vx − uy − vy
 
(ux + uy ) + (vx + vy )
=
(ux − uy ) + (vx − vy )
= T (u) + T (v)
 
cux + cuy
T (cu) =
cux − cuy
 
ux + uy
=c
ux − uy
= cT (u)

14
Exercise 5
Prove that the given transformation is a linear transformation, using the definition.
 
x  
x−y+z
T y =
 
2x + y − 3z
z
 
ux + vx
T (u + v) = T
uy + vy
 
(ux + vx ) − (uy + vy ) + (uz + vz )
=
2(ux + vx ) + (uy + vy ) − 3(uz + vz )
 
(ux − uy + uz ) + (vx − vy + vz )
=
(2ux + uy − 3uz ) + (2vx + vy − 3vz )
= T (u) + T (v)
 
cux − cuy + cuz
T (cu) =
c2ux + cuy − c3uz
 
ux − uy + uz
=c
2ux + uy − 3uz
= cT (u)

Exercise 7
Give a counterexample to show that the given transformation is not a linear transformation.
   
x y
T = 2
y x

    
3 6
T 2 =
3 36
 
3
6= 2T
3
 
6
6=
18

Exercise 9
Give a counterexample to show that the given transformation is not a linear transformation.
   
x xy
T =
y x+y

    
3 36
T 2 =
3 12
 
3
6= 2T
3
 
18
6=
12

15
Exercise 11
Find the standard matrix of the linear transformation.
   
x x+y
T =
y x−y
 
1 1
=
1 −1

Exercise 13
Find the standard matrix of the linear transformation.
 
x  
x − y + z
T y  =
2x + y − 3z
z
 
1 −1 1
=
2 1 −3

Exercise 40
Use matrices to prove the given statements about the transformations from R2 to R2 . If Rθ denotes a
rotation (about the origin) through the angle θ, then Rα ◦ Rβ = Rα + Rβ .
 
cos α − sin α
Rα =
sin α cos α
 
cos β − sin β
Rβ =
sin β cos β
  
cos α − sin α cos β − sin β
Rα ◦ Rβ =
sin α cos α sin β cos β
 
cos α cos β − sin α sin β − cos α sin β − sin α cos β
=
sin α cos β + cos α sin β − sin α sin β + cos α cos β
 
cos α cos β − sin α sin β −(cos α sin β + sin α cos β)
=
sin α cos β + cos α sin β cos α cos β − sin α sin β
 
cos(α + β) − sin(α + β)
=
sin(α + β) cos(α + β)
= Rα+β

Exercise 42
(a) If P is a projection, then P ◦ P = P .
 
d2x dx dy
d2x +d2y d2x +d2y
P = 
dx dy d2y

d2x +d2y d2x +d2y
 
d4x +d2x d2y d3x dy +dx d3y

P ◦P =  3 2x +d2y )23
(d (d2x +d2y )2 
dx dy +dx dy d4y +d2x d2y
(d2x +d2y )2 (d21 +d22 )2

=P

16
Exercise 44
Let T be a linear transformation from R2 to R2 . Prove that T maps a straight line to a straight line or a
point.
l = ~x + td~
~ = T (~x) + T (td)
T (~x + td) ~
~
= T (~x) + tT (d)
When t = 0, the result is a point, otherwise, the resulting mapping is a line.

Exercise 52
Prove that Pl (c~v ) = cPl (~v ) for an scalar c.
!
d~ · (c~v ) ~
Pl (c~v ) = d
d~ · d~
!
c(d~ · ~v ) ~
= d
d~ · d~
!
d~ · ~v ~
=c d
d~ · d~
= cPl (~v )

Exercise 53
Prove that T : Rn → Rm is a linear transformation if and only if:
T (c1 v~1 + c2 v~2 ) = c1 T (v~1 ) + c2 T (v~2 )

T (c1 v~1 + c2 v~2 ) = T (c1 v~1 ) + T (c2 v~2 )

= c1 T (v~1 ) + c2 T (v~2 )

Exercise 54
Prove that (as noted at the beginning of this section) the range of a linear transformation T : Rn → Rm
is the column space of its matrix [T ].
 
a11 . . . a1n
[T ] =  ... .. .. 

. . 
am1 . . . amn
 
u1
 .. 
~u =  . 
un
   
a11 a1n
T (~u) = u1  ...  + · · · + un  ... 
   
am1 amn
The range of T is a linear combination of the columns, so it is a subset of the column space. The converse
is also true, the column space is a subset of the range. Therefore, the two sets are equal.
If you have any questions, comments, or concerns, please contact me at [email protected]

17