1.

a. False
b. True
c. False
d. True
e. True
f. True
g. True

2.

3. Let B = { b1 , b1 , b1} = {1-3t2, 2+t-5t2, 1+2t} and

let C = {c1,c2,c3} = {1 , t , t2}.
The C coordinate vectors of b1, b2, and b3 are

1 2 1

[b1]c= 0 , [b2]c= 1 , [b3]c=  2 
   
     
 −3  −5  0 
  1 2 1
∴ p =  0 1 2
c← B  
 −3 −5 0 

Let x = t2, Then the coordinate vector [x]B satisfies

 0  1 2 1 0 1 0 0 3 
p [ x ]B  0  ⇒  0 1 2 0    0 1 0 −2 
[ x ]C =
=
     
C←B
1   −3 −5 0 1  0 0 0 1 

3
∴ [ x ]B  −2 
= ∴ t2 = 3(1-3t2) – 2(2+t-5t2) + (1+2t)
 
 1 

4.
a. The column vectors of matrix A are linear independent if the equation Ax = 0
has only the trivial solution.

 1 0 - 2 0  1 0 - 2 0 
- 2 1 6 0 ~ 0 1 2 0
   
 3 - 2 5 0 0 0 15 0

From the above augmented matrix, there are three basic variables and no free
variable. So, the equation Ax = 0 has only the trivial solution. Hence, the
column vectors of matrix A are linear independent.

− 4
b. T(u) = Au =  13  .
 
 23 

1 0 -2 −1 1 0 0 0.2  0.2 


c. Ax = b ⇒ -2 1 6 7  ~ 0 1 0 3.8 ∴ x =3.8 .
 
     
 3 -2 5 −4  0 0 1 0.6  0.6

0.2
d. No, x = 3.8 is the only one whose image under T is b.
 
0.6

 1 0 - 2 3  1 0 - 2 3 
e. Ax = c ⇒ - 2 1 6 2 ~ 0 1 2 8  . There is a solution for Ax = c.
   
 3 - 2 5 5 0 0 15 12
 Hence c is in the range of T.
5. Let y = β 0 + β1 x
Aβ = y
1 2  1 
1 3  β 1 
   0 =  
1 0   β1  0
   
1 −1 1 

The normal equation : ATA β = ATY

0.65
β = ( ATA)-1ATY =  
 0.1 
∴ y = 0.65 + 0.1x

0.85
0.95
= β 
Yˆ A= 
0.65
 
0.55
Y − Yˆ =
0.8062

6.
t p0 (t ) p1 (t ) p2 (t )
−3 1 −3 9
−1 1 −1 1
1 1 1 1
3 1 3 9

a. The orthogonal projection P̂2 of P2 onto the subspace spanned

By P0 and P1 is

< p2 , p0 > < p2 , p1 > 20 0

P̂2 = p0 + p1 = (1) + t = 5
< p0 , p0 > < p1 , p1 > 4 20

b. The vector q = p3 − pˆ 2 =t 2 − 5 will be orthogonal

to both P0 and P1 and { P0 , P1 ,q} will be an orthogonal basis for

span { P0 , P1 , P2 }. The vector of values for q at (-3,-1,1,3) is (4,-4,-4,4),

 so scaling by yields the new vector q =( 1 )(t2– 5).
4

 9 −9 
7. ATA=   The eigenvalues of ATA are 18 and 0.
 −9 9 
 1   1 
 2   2
λ1 = 18 v1 =   , λ2 = 0 v1 =  
− 1   1 
 2   2 

 1 1 
 2 2
V = [ v1 v2 ] =  
− 1 1 
 2 2 

 1 
3 2 0  3 
 
∑
1 − 2 
= 0 0 u1 = A v1 =
3 2  3
 0 0  2 

 3 

 1 
 3 
a. ∴ A = u1 D v2T = − 2  3 2  1 − 1 
 3  2 2 
 2 
 3 

 1 
 2  1 2 2
AT = v1 D-1 u1T =   1  −
− 1  3 2  3 3 3 
 2 

 1 
 6  1 2 2
=   3 −
− 1  3 3 
 6 

 1 1 1 
 18 −
9 9 
= 
− 1 1 1
− 
 18 9 9 
  1 1 1 
1   1 
 18 −
9 9  0 =  6 
b. xˆ = AT b =    
 
− 1 1 1
−  1  − 1 
 18 9 9   6 

 1   1 
 3  1   3 
1 2 2  
b
c. projColA = u1 u1 b =  − 2  
T
− 0 = − 2 
 3  3 3 3     3
 2  1   2 
 3   3 

 9 −4 4 
8. A =  −4 7 0  . The eigenvalues of A are 3, 9, and 15.
 
 4 0 11

 −2 
 3
 −2   
−2
λ1 = 3 , v 1 =  −2  , u1=  
   3
 1  1 
 
 3

 −1 
3
 −1  
, v2 =  2 
2
λ2 = 9 , u2=  
  3
 2  2
 
3

2
3
2  
−1
λ3 = 15 , v3 =  −1 , u3=  
  3
 2  2
 
3

 −2 −1 2 
 3 3 3 
  3 0 0 
−2
D = 0 9 0 
2 1
P = [u1 u2 u3] =  − and
 3 3 3  
 −2 0 0 15
−2 −2 
 
 3 3 3 
x = PY and the new quadratic form is
xTAx = 3y12 + 9y22 + 15y32