Math 102, Winter 2010,

Final Exam Review Problems, Part II

 
1 2 −1 3 1
 −2 0 −6 2 2 
(1) Let A = 
 4
 . If an echelon form of A is the matrix
1 10 −2 −3 
3 4 3 5 1
 
1 2 −1 3 1
 0 1 −2 2 1 
B=  0 0 1 0
, then:

0
0 0 0 0 0
1. Find a basis for Col(A), the column space of A,
 
a
 −2 
2. For what value of a is the vector u =  
 3  in the column space of A?
0
3. Find a basis for N ul(A), the null space of A,
4. Find rank(A), and nullity(A)
5. Find a basis for Row(A), the row space of A, consisting entirely of row
vectors from A
Solution :
1. Let A = [a1 a2 a3 a4 a5 ] , ai are columns of A. Pivot columns of A form
a basis for Col(A). Hence,
     

 1 2 −1  
   0   −6 
 −2     
  4  ,  1  ,  10 

 

3 4 3
is a basis of Col(A).
2. Since Cola(A) = Span {a1 , a2 , a3 }, the vector u belongs to Col(A) if
and only if the linear system [a1 a2 a3 ] x = u , where x = [x1 x2 x3 ]T
has at least one solution. Since
   
1 2 −1 | a 1 2 −1 | a
 −2 0 −6 | −2   0 1 −2 | − 1 + 1 a 
[a1 a2 a3 | u] =   
 4 1 10 | 3  =∼ · · · ∼  0 0 1 | − 1 − a
2 2 ,

2
1 1
3 4 3 | 0 0 0 0 | − 14 − 14 a
we get a = −1.
  
1 0 0 −1 −1
 0 1 0 2 1 
3. First note that the reduced echelon form of A is C = 
 0
.
0 1 0 0 
0 0 0 0 0
Therefore, the general solution of the linear system Ax = 0 in the para-
metric vector form is given by:
       
x1 s+t 1 1
 x2   −2s − t   −2   −1 
       
 
x =  x3  =   0     
 = s  0  + t  0  , s, t ∈ R.
 x4   s   1   0 
x5 t 0 1
   

 1 1  

   −1 
 −2   

Hence,   
 0  ,  0 





 1   0  

 
0 1
is a basis of N ul(A).
4. Rank(A) = dim(Col(A)) = 3, N ullity(A) = dim(N ul(A)) = 2.
5. Since Row(A) = Col(AT ) and
   
1 −2 4 3 1 −2 4 3
 2 0 1 4   0 4 −7 −2 
   
T 
A =  −1 −6 10 3  ∼ ··· ∼  0 0 0 2 
  ,
 3 2 −2 5   0 0 0 0 
1 2 −3 1 0 0 0 0
a basis for Row(A), the row space of A, consisting entirely of row vectors
from A is
{(1, 2, −1, 3, 1), (−2, 0, −6, 2, 2), (3, 4, 3, 5, 1)}.

(2)
½· ¸ ¾
a a+b
(a) Let W = | a, b ∈ R . Determine whether W is a subspace
3 b
of M2×2 , the vector space of all two by two matrices.
(b) Let v1 = (1, 2, 6), v2 = (3, 4, 1), v3 = (4, 3, 1), and v4 = (3, 3, 1). Deter-
mine whether these four vectors span R3 .
 Solution :
· ¸
0 0
(a) No, since zero vector 0 = does not belong to W .
0 0
(b) Let A = [v1 , v2 , v3 , v4 ]. Since an echelon form of A
 
1 3 4 3
 0 1 5 3 ,
2 2
0 0 392
17
2

has a leading entry in every row, vectors v1 , v2 , v3 and v4 span R3 .

(3)
(a) Find the standard matrix A = [T ], for the projection, T , of R2 onto the
line passing through the points (0, 0) and (1, 3).
(b) Is the linear transformation , T , in part (a) invertible? Justify your an-
swer.
Solution :
(a) Let a = (1, 3). The image under T of any vector u = (x1 , x2 ) from R3 is
given by:
µ ¶
u·a 1 3 3 9
T (u) = proja u = a = ( x 1 + x 2 , x 1 + x2 ).
||a||2 10 10 10 10
· 1 3 ¸
Hence, A = [T ] = 10 3
10
9 .
10 10

(b) Since det(A) = 0, the matrix A is not invertible. Hence, T is not invertible.
(4)
(a) Is the set {(1, 0, 0), (3, 2, 1), (1, 1, 0)} linearly independent? Show why or
why not.
(b) Let v1 = (−1, 1, 0, 0), v2 = (1, 2, 0, 1) and v3 = (−4, 1, 0, −1). Show that
the set of vectors {v1 , v2 , v3 } form a linearly dependent set in R4 . Ex-
press v1 as a linear combination of v2 and v3 .
Solution :
¯ ¯
¯ 1 1 3 ¯
¯ ¯
(a) Since ¯¯ 0 1 2 ¯¯ = 1 6= 0, the set is linearly independent.
¯ 0 0 1 ¯
(b) The vector equation
c1 v1 + c2 v2 + c3 v3 = 0
is equivalent to a linear system whose augmented matrix is [v1 v2 v3 | 0].
Since
   
−1 1 −4 | 0 1 0 3 | 0
 1 2 1 | 0   0 1 −1 | 0 
[v1 v2 v3 | 0] =   
 0 0 0 | 0  ∼ ··· ∼  0 0 0 | 0
,

0 1 −1 | 0 0 0 0 | 0

a nontrivial dependence relation among v1 , v2 , v3 is

(−3)v1 + v2 + v3 = 0.

Therefore,
1 1
v1 = v2 + v3 .
3 3

(5) Use the Gram-Schmidt process to find an orthonormal basis for the sub-
space W of R4 spanned by {u1 = (1, 1, 1, 1), u2 = (1, 0, 0, 1), u3 = (0, 2, 2, −1)}.
Solution : First note that {u1 , u2 , u3 } is a linearly independent set, thus
a basis for W .
v1 = u1 = (1, 1, 1, 1)
µ ¶
0 u2 · v 1 1 1 1 1
v 2 = u2 − 2
v1 = · · · = ( , − , − , )
||v1 || 2 2 2 2
Let v2 = 2v2 0 = (1, −1, −1, 1).
µ ¶ µ ¶
0 u3 · v 1 u3 · v 2 1 1 1 1
v 3 = u3 − 2
v1 − 2
v2 = · · · = ( , , − , − )
||v1 || ||v2 || 2 2 2 2

Let v3 = 2v3 0 = (1, 1, −1, −1). Then {v1 = (1, 1, 1, 1), v2 = (1, −1, −1, 1), v3 =
(1, 1, −1, −1)} is an orthogonal basis for the subspace W . By normalizing, we
get an orthonormal basis for W :
v1 v2 v3 1 1 1
{ , , } = { (1, 1, 1, 1), (1, −1, −1, 1), (1, 1, −1, −1)}.
||v1 || ||v2 || ||v3 || 2 2 2
(6) True or False? Give a brief explanation for your answer.
(a) If AT A = A, the A is symmetric.
(b) If the reduced row-echelon form of A has a row of zeros, then det(A) = 0.
(c) det(I + A) = 1 + det(A).
(d) If u 6= 0 and u × v = u × w, then v = w.
(e) It is possible to have proja u = proju a.
Solution :
(a) True, since AT = (AT A)T = AT (AT )T = AT A = A.
(b) True, since A is not invertible
(c) False, since in general
· det(A
¸ + B) 6= det(A) + det(B).. Or, taking I to be
1 2
2 × 2 and A = we get det(I + A) = 4 and 1 + det(A) = −1.
3 4
(d) False, since for v 6= w we have u = v − w 6= 0 and u × v = u × w.
(e) True, u = a.

(7) Let W = Span{v1 = (2, 2, −1, 0, 1), v2 = (−1, −1, 2, −3, 1), v3 = (1, 1, −2, 0, −1), v4 =
(0, 0, 1, 1, 1)}, be a subspace of R4 .
(a) What is the dimension of W ?
(b) Is (1, 2, 0, 0, 0) in W ?
(c) Find a basis for W ⊥ , the orthogonal complement of W .
Solution :
(a) If A denotes the matrix whose columns are vectors v1 = (2, 2, −1, 0, 1), v2 =
(−1, −1, 2, −3, 1), v3 = (1, 1, −2, 0, −1) and v4 = (0, 0, 1, 1, 1), then W =
Col(A). Since
   
2 −1 1 0 1 1 −1 1
 2 −1 1 0   0 −3 3 −2 
   
A =  −1 2 −2 1  ∼ · · · ∼ 
 
 0 0 1 −1  ,
 0 −3 0 1   0 0 0 0 
1 1 −1 1 0 0 0 0
a basis for W is {v1 = (2, 2, −1, 0, 1), v2 = (−1, −1, 2, −3, 1), v3 =
(1, 1, −2, 0, −1) and dim(W ) = 3.
(b) Let u = (1, 2, 0, 0, 0). Since an echelon form of the augmented matrix
[v1 v2 v3 | u] has a row of the form [0 0 0 | 1], vector u is not in W .

(c) Let A be the matrix from part (a). Then , the orthogonal complement of
W is: W ⊥ = (Col(A))⊥ = N ull(AT ).
   
2 2 −1 0 1 1 1 0 0 0
 −1 −1 2 −3 1   0 
[AT | 0] =   ∼ ··· ∼  0 0 1 0 .
 1 1 −2 0 −1   0 0 0 1 0 
0 0 1 1 1 0 0 0 0 1

Hence, the general solution of AT x = 0 is given by x = s(−1, 1, 0, 0, 0),

that is {(−1, 1, 0, 0, 0)} is a basis for W ⊥ .

(8) Let V be the vector space of all polynomials of degree at most 2 equipped
with the inner product defined by

hp, qi = p(−1)q(−1) + p(0)q(0) + p(1)q(1), p(x), q(x) ∈ V.

Find a nonzero polynomial in V which is orthogonal to both p(x) = 1 + x + x2

and q(x) = 1 − 2x + x2 .
Solution : Let r(x) = ax2 + bx + c be a polynomial we are looking for.
We need to find a, b and c so that hr, pi = 0 and hr, qi = 0. Thus, we need a
nontrivial solution of the linear system:

4a + 2b + 5c = 0

4a − 4b + 5c = 0.
Clear, b = 0 and a = − 54 c where c is any nonzero real number. For example if
c = 4 we get a = −5 and r(x) = −5x2 + 4.