Math 2660 Topics in Linear Algebra, Key

3.2
1a,b,c,d,2a,b,c,3a,b,c,d,4a,b,9a,c,10a,b,e,11,12

1 (a) S = {x = (x1 , x2 )T : x1 + x2 = 0}. (i) 0 ∈ S since 0 + 0 = 0 so S is not empty. (ii) If

x, y ∈ S, i.e, x1 + x2 = 0 = y1 + y2 , then x + y ∈ S since x + y = (x1 + y1 , x2 + y2 )T
and x1 + y1 + x2 + y2 = x1 + x2 + y1 + y2 = 0. (iii) If x ∈ S and α ∈ R, i.e, x1 + x2 = 0,
then αx = (αx1 , αx2 ) ∈ S since αx1 + αx2 = α(x1 + x2 ) = α · 0 = 0.
So S is a subspace of R2 .
(b) S = {x = (x1 , x2 )T : x1 x2 = 0} is not a subspace of R2 since (1, 0)T , (0, 1)T ∈ S but their
sum (1, 1)T 6∈ S. Remark: S is closed under scalar multiplication and S is nonempty
since 0 ∈ S.
(c) S = {x = (x1 , x2 )T : x1 = 3x2 }. (i) 0 ∈ S since 0 = 3 · 0 so S is not empty. (ii) If
x, y ∈ S, i.e, x1 = 3x2 , y1 = 3y2 , then x + y ∈ S since x + y = (x1 + y1 , x2 + y2 )T
and x1 + y1 = 3x2 + 3y2 = 3(x2 + y2 ). (iii) If x ∈ S and α ∈ R, i.e, x1 = 3x2 , then
αx = (αx1 , αx2 ) ∈ S since αx1 = α(3x2 ) = 3(αx2 ).
So S is a subspace of R2 .
(d) S = {x = (x1 , x2 )T : |x1 | = |x2 |} is not a subspace of R2 since (1, −1)T , (1, 1)T ∈ S but
their sum (2, 0)T 6∈ S as |2| = 2 6= 0 = |0|.

2 (a) S = {x = (x1 , x2 , x3 )T : x1 +x3 = 1} is not a subspace of R3 since (1, 0, 0)T , (0, 0, 1)T ∈ S
but their sum (1, 0, 1)T 6∈ S as 1 + 1 = 2 6= 1.
(b) S = {x = (x1 , x2 )T : x1 = x2 = x3 }. (i) 0 ∈ S since 0 = 0 = 0 so S is not empty.
(ii) If x, y ∈ S, i.e, x1 = x2 = x3 , y1 = y2 = y3 , then x + y ∈ S since x + y =
(x1 + y1 , x2 + y2 , x3 + y3 )T and x1 + y1 = x2 + y2 = x3 + y3 . (iii) If x ∈ S and α ∈ R,
i.e, x1 = x2 = x3 , then αx = (αx1 , αx2 , αx3 ) ∈ S since αx1 = αx2 = αx3 .
So S is a subspace of R3 .
(c) S = {x = (x1 , x2 , x3 )T : x3 = x1 + x2 }. (i) 0 ∈ S since 0 = 0 + 0 so S is not empty.
(ii) If x, y ∈ S, i.e, x3 = x1 + x2 , y3 = y1 + y2 , then x + y ∈ S since x + y =
(x1 + y1 , x2 + y2 , x3 + y3 )T and x3 + y3 = (x1 + x2 ) + (y1 + y2 ) = (x1 + y1 ) + (x2 + y2 ).
(iii) If x ∈ S and α ∈ R, i.e, x3 = x1 + x2 , then αx = (αx1 , αx2 , αx3 ) ∈ S since
αx3 = α(x1 + x2 ) = αx1 + αx2 .
So S is a subspace of R2 .

3 (a) Let S be the set of all 2 × 2 diagonal matrices. (i) The zero matrix is in S. (ii) The sum
of two diagonal matrices is still a diagonal matrix. (iii) The scalar multiple of a diagonal
matrix is still a diagonal matrix.
So S is a subspace of R2×2 .
(b) The set of 2 × 2 triangular matrices is a subspace of R2×2 and it is similar to (a).
(c) Similar to (b).
(d) Let S be the set of all 2 × 2 matrices with a12 = 1. (i) The zero matrix is not in S. So it
is not a subspace of R2×2 (if it were, then it would contain the zero matrix (the identity
element of R2×2 )).

1
 4 (a) " # " #
2 1 0 2 1 0
[A | 0] =
3 2 0 R2 − 32 R1 0 21 0

so the null space of A is {(0, 0)T }.

(b) " # " #
1 2 −3 −1 0 1 2 −3 −1 0
[A | 0] =
−2 −4 6 3 0 R2 + 2R1 0 0 0 1 0

so the null space of A is {(−2α + 3β, α, β, 0)T : α, β ∈ R}.

9 (a) Set α(2, 1)T + β(3, 2)T = (a, b)T ∈ R2 . Consider the augmented matrix
" # " #
2 3 a 1 3 a
1 2 b R2 − 21 R1 0 21 b − 21 a

The system is consistent (β = 2b − a and α = a − 3(2b − a) = 4a − 6b). So the set is a

spanning set.
(c) Set α(−2, 1)T + β(1, 3)T + γ(2, 4)T = (a, b)T ∈ R2 . Consider the augmented matrix
" # " #
−2 1 2 1 −2 1 2 1
1 3 4 b R2 + 12 R1 0 52 5 b + 21

The system is consistent so that the set is a spanning set.

10 (a) Consider the augmented matrix

     
1 0 1 a 1 0 1 a R1 − R3 1 0 0 a−c+b
     
 0 1 0 b   0 1 0 b   0 1 0 b .
0 1 1 c R3 − R2 0 0 1 c−b 0 0 1 c−b

In other words, the system is consistent so that it is a spanning set for R3 . Indeed
(a, b, c)T = (a − c + b)(1, 0, 0)T + b(0, 1, 1)T + (c − b)(1, 0, 1)T .
(b) Consider the augmented matrix
     
1 0 1 1 a 1 0 1 1 a R1 − R 3 1 0 0 0 a−c+b
     
 0 1 0 2 b   0 1 0 2 b   0 1 0 2 b .
0 1 1 3 c R3 − R2 0 0 1 1 c−b 0 0 1 1 c−b

It has infinitely many solution, but the point is that it is consistent. So it is a spanning
set for R3 .
(e) Consider the augmented matrix
     
1 0 a 1 0 a 1 0 a
     
 1 2 b  R 2 − R1  0 2 b−a   0 2 b−a .
3 1 c R3 − 3R1 0 1 c − 3a R2 − 21 R2 1 1
0 0 c − 3a − 2 b + 2 a

It has no solution if c − 52 a − 12 b 6= 0, say (a, b, c)T = (0, 0, 1)T . So it is not a spanning

set for R3 .

2
11 (a) x ∈ span {x1 , x2 } means that x = αx1 +βx2 for some α, β ∈ R. Consider the augmented
matrix    
−1 3 2 −1 3 2
   
[x1 x2 |x] =  2 4 6  R2 + 2R1  0 10 10  R2 /10
3 2 6 R3 + 3R1 0 11 13
   
−1 3 2 −1 3 2
   
 0 1 1   0 1 1 
0 11 13 R3 − 11R2 0 0 2
It is inconsistent. So x 6∈ span {x1 , x2 }.
(b) y ∈ span {x1 , x2 } means that y = αx1 +βx2 for some α, β ∈ R. Consider the augmented
matrix
   
−1 3 −9 −1 3 −9
   
[x1 x2 |y] =  2 4 −2  R2 + 2R1  0 10 −20  R2 /10
3 2 5 R3 + 3R1 0 11 −22
   
−1 3 −9 −1 3 −9
   
 0 1 −2   0 1 −2 
0 11 −22 R3 − 11R2 0 0 0
It is consistent and α = 3 and β = −2. So y ∈ span {x1 , x2 }.

12 Let {x1 , . . . , xn } be a spanning set for a vector space V .

(a) If we add an additional vector xk+1 to the spanning set, we still have a spanning set.
It it because each x ∈ V can be written as a linear combination of x1 , . . . , xk , say,
x = α1 x1 + · · · + αk xk so that we have x = α1 x1 + · · · + αk xk + 0 · xk+1 .
(b) In general it is not true, for example e1 , e2 span R2 but e1 does not span R2 . However
it can be true for some situations, for example e1 , e2 , e1 + e2 span R2 (by (a)) and if
we remove e1 + e2 , we still have a spanning set.