Review Problems for Linear Algebra Some answers are posted at the end.

Updated April 2,2020

1. (1) For any matrix A, the system of linear equations Ax = 0 always has at least one
solution. T F

(2) If Ax = b has many solutions for a certain b, then Ax = 0 must have many
solutions too. T F

(3) If Ax = 0 has many solutions, then Ax = b has many solutions for any b. T F

(4) If the reduced row echelon form of a coefficient matrix has a zero row, then the
system of equations is inconsistent. T F

(5) If a linear system of equations has fewer equations than variables, then the system
has many solutions. T F

(6) If the rows of a matrix A are linearly independent, then Ax = b always has a
unique solution. T F

(7) If line l has direction v and plane P has normal n, then if v · n = 0, l does not
intersect P . T F

(8) If n1 and n2 are normals of the planes P1 and P2 respectively, and if n1 · n2 6= 0,

then P1 and P2 intersect. T F

(9) For any vectors u, v of the same dimension, ku + vk < kuk + kvk. T F

(10) For any vectors u, v of the same dimension, |u · v| < kuk kvk. T F

(11) If A and B are invertible n × n matrices, then A + B is an invertible matrix. T F

(12) If A is an invertible matrix, then so is AT . T F

(13) For any two matrices A, B of the same size, (A − 2B)T = AT − 2B T . T F

(14) For any two n × n matrices A, B, A2 − B 2 = (A − B)(A + B). T F

(15) Every elementary matrix has determinant equal to 1. T F

(16) If det(AB) = 0, then at least one of A or B is singular. T F

(17) Every n × n matrix can be expressed as a product of elementary matrices. T F

(18) If Ax = b has a many solutions, then the columns of A must be linearly dependent. T F

(19) If w ∈ span{v1 , . . . , vk }, then w can be written as a linear combination of v1 , . . . , vk

in only one way. T F

1
 (20) If w ∈
/ span{v1 , . . . , vk }, then then {w, v1 , . . . , vk } is linearly independent. T F

(21) If W = span{v1 , . . . , vk }, then dim(W ) = k. T F

(22) If u · v = 0, then ku − vk2 = kuk2 + kvk2 . T F

(23) If u · v = 0 for some v 6= 0, u 6= 0, then u × v 6= 0 T F

(24) If v × u = 0 for some v 6= 0, u 6= 0, then u · v 6= 0 T F

(25) For any matrix A, col(A)⊥ = null(AT ). T F

(26) Every n × n matrix has a (real) eigenvalue. T F

(27) If v is an eigenvector of A, then v is an eigenvector of Am for every integer m ≥ 1. T F

(28) For any matrix M, M M T and M T M are square matrices. T F

(29) For any matrix M , if M T M is invertible, then M (M T M )−1 M T = I. T F

(30) For any matrix A, rank(A) = rank(AT ). T F

(31) If A and B are matrices of the same size, then rank(A + B) = rank(A)+ rank(B). T F

(32) If T and S are linear transformations Rn → Rn , then [T + S] = [T ] + [S]. T F

2. Check whether each of the set of numbers (20, −10, 9, −15, −14),
(31, −15, 25, 3, 3), (30, −16, 28, 3, 6) are solutions of the following system of equations.

x1 + 2x2 − x4 + x5 = 1
3x2 + 2x3 − x5 = 2
3x2 + x3 + 7x4 = 1
3. Here is a row reduced augmented matrix of a system of equations;
1 2 0 0 1 0 1
 

 0 0 1 0 −2 1 0 

0 0 0 1 1 0 −3
 
 
0 0 0 0 0 0 0

(i) Write out the general solution of the system (i.e., the solution set).
(ii) Write out 3 solutions of the system.
(iii) Is there a positive solution? That is, a solution where all the variables are positive?
4. Reduce the following system of equations to row echelon form in order to answer ques-
tions below. Notice there is ’c’ in the equations....
x + y − z = 1
2x + 3y + c z = 3
x + c y + 3z = 2

Determine c, if possible, so that there is a unique solution, no solution, or many solutions.

2
 5. Determine all values of the parameters k, p, if there are any, such that the following
system of equations has no solution, a unique solution or many solutions. In the case
where there are solutions (one or many) write out the solutions.

2x + ky = p
x + y = 1

6. Find the parametric equation of the line in R2 (that is, of the form x = x0 + t v) that
passes through the point (2, 3) and is parallel to v = (−1, 2). Sketch this line. Find
some points on this line. Is the point (1, −2) on this line? Is the point (−2, 11) on this
line?

7. Express the line y = 2x − 3 in the form x = x0 + t v.

8. Express the line x = (2, −3) + t(−2, 2) in the form y = mx + b.

9. Find the parametric equation of the line in R3 that passes through the points P1 (1, 2, 3)
and P2 (−2, 1, 1). Find some points on this line. Is the point (2, 0, 1) on this line?

10. Where does the line r = (1, −1) + t(1, 1) intersect the line q = (2, 2) + s(3, 1)? Sketch
these lines and verify your answer. (Do this by solving a system of equations for the two
unknowns t and s.)

11. (a) Set up the system of equations that will find the intersection between these two
lines in R2 ;

x1 = (1, 1) + t(2, −4)

x2 = (0, 2) + s(1, 2)

(b) Set up the system of equations that will find the intersection between these two
lines in R3 ;

x1 = (1, 1, 1) + t(2, −4, 4)

x2 = (0, 2, −1) + s(−1, 2, −3)

12. Set up the system of equations that will determine the intersection of three lines in R2 .
From the size of this system, deduce that it is not likely that they will intersect.

13. Show that the augmented matrix you obtain when determining the intersection point of
the two lines in Rn , q(t) = qo + t v, r(s) = ro + s w, or in coordinates,

q(t) = (q1 , q2 , . . . , qn ) + t(v1 , v2 , . . . , vn ), r(s) = (r1 , r2 , . . . , rn ) + s(w1 , w2 , . . . , wn )

3
 is;
v1 −w1 r1 − q1
 

 v2 −w2 r2 − q2 

 .. .. ..  (3 columns here)
. . .
 
 
vn −wn rn − qn

14. Do the following pairs of equations describe the same line? (Answer yes or no for each
of the 3 pairs.)

x1 = (1, 1, 1) + t(2, −4, 4)

x2 = (0, 2, −1) + s(−1, 2, −2)

x1 = (1, 1, 1) + t(2, −4, 4)

1
x2 = (2, −1, 3) + s(− , 1, −1)
2

x1 = (1, 1, 1) + t(2, −4, 4)

x2 = (0, 2, −1) + s(1, 2, −2)

15. Find the equation of the plane that is perpendicular to the line r = (1, −1, 1) + t(1, 1, 2)
and that passes through the point (1, 2, 3). Find some points on this plane. Is the point
(2, −2, 3) on this plane?
figures);

16. Find the intersection between the planes −3x − 4y + 5z = 2 and 2x − 3y + z = 4.

Express this intersection parametrically (i.e., show that it is a line), and find the angle
of intersection between these planes. Verify your answer (i.e., show that this line lies in
both planes).

17. Find the intersection between the planes −3x − 4y + 5z = 2, 2x − 3y + z = 4, and

−4x + 2y + z = 2.

We solve the system;

−3x − 4y + 5z = 2
2x − 3y + z = 4
−4x + 2y + z = 2

The unique solution is (x, y, z) = (−68, −82, −106). We expect a system of 3 equations
in 3 unknowns to usually have a unique solution; this is an algebraic reason why we
expect 3 planes in R3 to intersect in a point (we also expect that from geometrical
reasoning; two planes intersect in a line, and a line and plane intersect at a point.)

18. Explain why it is unlikely that 4 or more planes in R3 will intersect.

4
19. Find the algebraic equation of the plane that passes through the origin, is parallel to
the line x = (−1, 0, 2) + t(2, −1, 1), and contains the point (1, 2, 3).

20. Consider the plane with equation

n · (x − uo ) = 0, n = (3, −2, 1), uo = (a, 2, 3)

and the line with equation

x = xo + tv, xo = (1, 1, 1), v = (1, −1, a)

Determine all values of a such that this line intersects the plane. Find these points of
intersection.

21. (a) Find all planes that contain the two points (2, −1, 3), (1, 0, 1).

(b) Find all planes that contain these two points and are parallel to the line x(t) =
(2, 2, −3) + t(−1, 1, 2).

22. Find the plane that contains the three points (1, 1, 1), (2, 0, 2), (−3, 2, 1) in two ways; (i)
solve a system of linear equations, and (ii) use the cross product.

23. (a) Let v = (−1, −2, 3). Find a vector of magnitude 4 in the opposite direction as v.
(b) Determine k, if possible, such that the angle between the vector u = (1, k, 2) and
v (same v as in part (a) ) is 60 degrees.

24. Find the flow in all the arcs of the following network.

25. Find the cubic polynomial that passes through the points P1 (−2, 2), P2 (1, 3), P3 (0, 2), P4 (4, −2).
Sketch these points and the curve.

26. Find all second order polynomials that pass through the points P1 (−2, 1), P2 (1, 3).

27. Balance the following chemical equation;

C3 H8 + O2 → CO2 + H2 O

5
28. Using only the ingredients potato, rice, corn, beans, find a diet (mixture of ingredients)
that satisfies the following requirements;
contains 40g of carbohydrates, 25g of protein, 15g of fat.
The nutrition composition of each ingredient, per 100g of that ingredient, is
grams of potato rice corn beans
carbo 12 8 5 15
protein 4 6 5 15
fat 4 2 5 6
Formulate a mathematical model (system of linear equations) whose solution will tell
you how much of each ingredient to use.
2 2 −1 1
 
 0 0 0 0 
29. Let C =  .

Explain why C is non-invertible (is singular) without
 −2 1 7 2


4 0 3 2
doing any calculations.

30. Suppose A, B, C are invertible matrices, and all the following products are defined.
Simplify the expressions (remove all brackets);

(A−1 BC −1 )−1
((B T C −1 )T )−1

31. Let A, B, C be n × n invertible matrices. Solve the following expression for the matrix
H (i.e., express H in terms of A, B, C).
 −1
HAT + B = −2C T B −1

32. Let
     
2 −1 3 −1 0 1 3 1 1
A =  −2 0 2 
, B= 3 4 −1 
, C= 4 1 −1 
  

5 7 2 2 0 1 2 1 −1

Find a matrix Y , if possible, that satisfies the equation

(2tr(A)Y )T − 3B = 4C

33. Determine the elementary row operations that these elementary matrices perform;

1 0 0 0
 

 −2 1 0 0 

0 0 1 0
 
 
0 0 0 1

6
 1 0 0 0
 

 0 1 0 0 

0 0 0 1
 
 
0 0 1 0
1 0 0 0
 

 0 −1 0 0 

0 0 1 0
 
 
0 0 0 1
1 0 0 0
 

 0 1 0 0 

0 −5 1 0
 
 
0 0 0 1

34. The matrix B below was obtained by applying several elementary row operations to the
matrix A. Find a matrix C such that B = CA. Verify your answer (i.e., perform the
multiplication CA).
   
0 1 3 −2 −1 4
A= 2

1 −4 
 −→ B= 0

1 3 

2 3 2 0 2 6

35. Write out the 5×5 matrix E that performs the following three elementary row operations
on a matrix when it multiplies that matrix on the left; interchange rows 2 and 4, then
multiply row 3 by 7, then subtract 4 times row 1 from row 5 (in that order!).

Now find a 5 × 5 matrix F that performs the following column operations on a matrix
A when you form the product AF ; interchange columns 2 and 3 and then subtract 2
times column 4 from column 5.

36.  
1 0 2
A=

2 −1 3 

4 1 8
Write A and A−1 as a product of elementary matrices.

 
2 3 −2 5 1
37. Let A =  3 −1

2 0 4 
 and consider the problem Ax = b. Determine
4 −5 6 −5 7
conditions on b = (b1 , b2 , b3 ) that would guarantee the system is consistent.
Express your answer also as a sum of vectors with parameters. Use this to find a basis
for the range of A.

7
38. Determine whether the following subsets of vectors are subspaces. If they are not,
explain why not.

• All vectors in R4 of the form (a, 1, b, c), a, b, c ∈ R.

• {v = (0, a, b, a − b) | a, b ∈ R}
" #
a c
• All 2 × 2 matrices of the form , a, b, c ∈ R. (Hint: Think of each 2 × 2 matrix
c b
" #
a b
as a vector (a, b, c, d) ∈ R4 .)
c d
 
2 −3 0 1
T
39. Let A =  5 −1 −4 2 
 , b = (1, 2, 3).


−1 0 0 3
Solve the homogeneous system Ax = 0 by row reduction.
Verify that (0, 0, 0, 1)T is a solution of the non-homogeneous system Ax = b.
Without doing any more calculations, write out the general solution of the non-homogeneous
system (that is, write out all the solutions to the non-homogeneous system).
Now solve the non-homogeneous system by row reduction. Show that the general solu-
tion you find is the same as the general solution you found above.

40. Determine whether the vector w is a linear combination of the vectors {v1 , v2 , v3 }.

w = (1, −2, 5); v1 = (1, 1, 1), v2 = (1, 2, 3), v3 = (2, −1, 1)

w = (2, −5, 3); v1 = (1, −3, 2), v2 = (2, −4, −1), v3 = (1, −5, 7)
If so, write out the linear combination and verify.

41. Determine conditions on a, b, c so that the vector w = (a, b, c) is in

span{(2, 1, 0), (1, −1, 2), (0, 3, −4)}. Use this result to find a basis for
span{(2, 1, 0), (1, −1, 2), (0, 3, −4)}.

42. Determine whether the following set of vectors is linearly independent or not. If not,
find a maximal linearly independent set (i.e., find the most vectors that form a linearly
independent subset).

(1, −2, 5, −3), (2, 3, 1, −4), (0, 0, 0, 10), (3, 8, −3, 5)

Do this 2 ways: (1) Put these vectors into a matrix as columns and row reduce (use the
Pivot Theorem), and (2) Put these vectors into a matrix as rows and row reduce.

8
43. Find a maximal linearly independent subset of vectors from the set S = {u1 , u2 , u3 , u4 };

u1 = (1, 1, 3, 4)
u2 = (2, 7, 8, −4)
u3 = (2, 3, 5, −1)
u4 = (−1, 2, 1, 2)

and (b) express the remaining vectors as a linear combination of the linearly independent
ones. For your reference we provide here two matrices and their reduced row echelon
forms;

1 1 3 4 1 0 0 −41/7
   
 2 7 8 −4   0 1 0 −30/7 

−→
  
2 3 5 −1  0 0 1 33/7 
   
 
−1 2 1 2 0 0 0 0
1 2 2 −1 1 0 0 1
   
 1 7 3 2   0 1 0 1 
 
−→  
3 8 5 1  0 0 1 −2
   
  
4 −4 −1 2 0 0 0 0

44. The reduced row echelon form of the matrix A is R as given here via E;

1 0 1 −1 1 0 0 −2
   
 2 5 0 6   0 1 0 2 
A=  −→ R = EA = 
   
3 −3 7 −5  0 0 1 1 

 
4 1 −2 −8 0 0 0 0

5 −0.5 0 2.5
 

1  −2 3 0 −1 
E=  
9 0.5 0 −2.5
 
14  
−84 7 14 7
Use this to answer the questions below.

(a) Is the set of vectors

S = {(1, 2, 3, 4), (0, 5, −3, 1), (1, 0, 7 − 2), (−1, 6, −5, −8)}

linearly dependent or linearly independent? Your answer should make reference to the
definition of linear independence. What is dim(span(S))?

The matrix A has these vectors as its columns. By the Pivot Theorem, we see that the
first 3 columns of A form a maximally linearly independent set. In particular, the set S
is linearly dependent, and dim(span(S)) = 3.

(b) Is the vector v = (1, 1, 1, 1) in span(S)? Explain.

9
 (c) Find all vectors b such that b ∈ span(S). Express your answer as a linear combi-
nation of linearly independent vectors (i.e., in the form b = t1 v1 + · · ·).

45. Find a basis for span{(1, −3, 2), (2, −4, −1), (1, −5, 7)} and determine the dimension of
this subspace.

46. Show that for any square matrix A, (A + AT ) is symmetric and (A − AT ) is skew-
symmetric.

47. Show that if D is diagonal, L lower triangular, and U upper triangular, then (D ± L) is
lower triangular and (D ± U ) is upper triangular.

48. Show that if D is a diagonal matrix, then so is Dn for any n ≥ 1.

49. Let
−4 2 6 2 2 0 0 0 −2 1 3 1
     
 2 3 3 7   −1 2 0 0   0 2 3 4 
A= ,

L= ,

U = 
−4 2 0 5 2 0 3 0 0 0 −2 1
   
     
−2 −1 0 1 1 −1 0 2 0 0 0 2

Show that A = LU and use this to solve the system Ax = b where b = [3 2 − 1 1]T .

50. Compute the determinant of the following matrix. First, by expansion across the first
row, then across the second row, and down the second column. Verify that these all give
the same answer.

51. Compute the determinant of the following matrix anyway you want (i.e., expansion down
any column or row, plus possibly applying elementary row operations first).
 
5 −2 3 1 3
2 0 1 0 −2 
 

 

 0 1 3 2 2 


 2 2 1 1 −2 

−2 1 3 1 2

52. If det(A) = 5, det(B) = −3 what is det(A2 B)− det(3B) + 2 det(A−2 )?

a b c
53. If d e f = −2, compute, if possible, the following determinants;
g h i

a − 2d + g b − 2e + h c − 2f + i c + f −b − e a + d a 2b c
−3d −3e −3f , f −e d , 2d 2e 2f
g−d h−e i−f i + 2c −h − 2b g + 2a g + 3a h + 3b i + 3c

10
54. Prove that if A is a square matrix, then det A 6= 0 if and only if A is invertible.

55. Find the adjoint of the matrix

−1 1 2 0
 

 2 0 1 3 

1 −1 0 3
 
 
4 1 0 1
Verify your answer using the formula
1
adj(A) = A−1
det(A)

56. Determine x3 in the solution of Ax = b where A is given above, b = [2 − 1 3 2]T ,

using Cramer’s Rule.
57. Find a basis for a⊥ where a = (−1, 1, 2, 4, 5).
58. Determine the area of the parallelogram which has corners at (3, 2), (4, 5), (2, 4). Begin
by making a sketch.
59. Determine the volume of the parallelepiped which has corners at
(1, 1, 1), (2, 3, 2), (0, −2, 1), (2, 4, 0). Begin by making a sketch.
60. Compute w = u × v where u = (3, −2, 1), v = (1, −3, 4) and sketch all three vectors.
Verify that w is orthogonal to both u and v.
61. Let z = 3 − 2i, w = −4 + 5i. Find 2zw, w2 , w̄z, 3z/w, |zw|.
62. Express z (given above) in polar form, and from this calculate z 9 , and all the cube roots
of z; z 1/3 . Sketch z, z 9 , z 1/3 together on the complex plane and identify their moduli
and principle arguments.
63. Find all the roots of 2x2 − 3x + 4.
 
1 −3 3
64. Find all the eigenvectors of A =  3 −5 3 
.


6 −6 4
Find a basis for each eigenspace.
What are the algebraic and geometric multiplicities of the eigenvalues you found for A?
 
−3 1 −1
65. Let B =  −7 5 −1 .
 

−6 6 −2
Without doing much calculation (i.e., without calculating any determinants or doing
any row reduction), determine whether any of the following vectors are eigenvectors of
B. If a vector is an eigenvector, identify the eigenvalue.
       
1 2 1 0
 1 ,  2 ,  −1  ,  1 
       

1 0 1 1

11
66. Find all three roots of p(x) = 2x3 + 10x2 + 6x − 2. (Begin by guessing one root, factor
that out and then find the roots of the remaining quadratic). Write out p(x) in factored
form, i.e., in the form p(x) = a(x − a1 )(x − a2 )(x − a3 ).

67. What are the possible algebraic and geometric multiplicities of an eigenvalue of a 5 × 5
matrix?

68. (a) If T : Rn → Rn is orthogonal projection onto the subspace W , does T (which means,
[T ]) have any eigenvectors? If so, what are the eigenvalues for these eigenvectors? Does
T necessarily have to have any (real) eigenvalues? Can you say anything about det([T ])?
(b) Suppose T is reflection across a plane in R3 . Does T have any eigenvectors? Can
you say anything about det(T ) (= det([T ])?

69. Decide which of the following 4 functions are linear transformations or not. If it is, find
the standard matrix for it, and determine whether it is one-to-one, and whether it is
onto.
F : R3 → R3 ; F (x, y, z) = (2x − y + 1, x − 4z, 7y − z)
T : R3 → R4 ; (4x1 − 2x2 , 3x1 x3 , x1 − x2 , 2x2 − x3 )
T (x1 , x2 , x3 ) = (4x3 − x1 , x1 + 2x3 , 3x2 − 2x1 + x3 )
G : R3 → R2 ; G(x) is the vector in R2 that is obtained by first projecting x ∈ R3
onto the xy − plane (this is R2 ), and then rotating that vector by 45o counter clockwise
(when looking down the z-axis at the xy-plane). Why is this a linear transformation?
Find the matrix [G].

70. Show that the following matrix;

√1 √1 √1
 
3 2 6
√1 − √12 √1
 
A=
 3 6


√1 0 − √26
3

is an orthogonal matrix. Calculate its determinant and its inverse.

Show that the following matrix is not orthogonal;
 
√1 0 √1
 2 2 
B=
 0 −1 0 

√1 − √1 − √1
3 3 3

71. Circle all the matrices below that are orthogonal matrices.
 

0 1 0 0

√ √ √ 
1 0 1 0 3/√11 −1/√6 −1/√66
 1 0 1 0 
(a)  0 1 0 0  (b) (c)  1/√11 2/√6 −4/√66 
     
0 0 0 1
 
 
0 0 0 1 1/ 11 1/ 6 7/ 66
0 0 1 0

12
    √ √ √ 
− cos θ sin θ 0 1/ 3 −1/√3 1/√3
(d)  − sin θ − cos θ 0  (e)  √0 1/√2 −1/√2 
  

0 0 −1 2/ 6 −1/ 6 −1/ 6

72. Why is rotation around the z-axis followed by reflection across the yz-plane an orthog-
onal transformation in R3 ? Why are shears not orthogonal transformations?
73. Determine whether the following transformations are one-to-one and whether they are
onto. And to do this, determine the kernel and the range. (ker(A)=null(A); Ax = 0,
range(A); Ax = b)

(1) Projection in R3 onto the yz-plane.

 
1 −8 −7 −4
(2) A = 

2 −3 −1 5 

3 2 5 14
 
1 0 1 4
(Hint: the reduced row echelon form of A is R =  0 1 1 1  via
 

0 0 0 0
 
−3 8 0
1 
E=  −2 1 0   ; R = EA )
13
13 −26 13
(3) T (x, y, z) = (−x + 3y + 2z, 2x + 4z, x + 3y + 6z)
74. Prove that ker(A) = 0 ⇐⇒ A is one-to-one.
75. For each of the following pairs of transformations determine whether the compositions
T1 ◦ T2 and T2 ◦ T1 are defined;
T1 : R3 → R3 , T2 : R3 → R3
T1 : R5 → R3 , T2 : R2 → R5

76. (a) Find the standard matrices of the transformations T1 , T2 and verify the composition
property [T2 ◦ T1 ] = [T2 ][T1 ];
T1 : R3 → R3 ; rotation by 30o about the y-axis
T2 : R3 → R2 ; T (x, y, z) = (y, x)
That is, first describe what the transformation T2 ◦ T1 does, and then verify with the
matrix [T2 ][T1 ]. Find [T2 ◦ T1 ].
(b) What is m and n in the formula T2 ◦ T1 : Rn → Rm ? Is T1 ◦ T2 defined?
(c) Find the kernel and range of T2 ◦ T1 . Is T2 ◦ T1 one-to-one? Onto?
77. Let W = span {v1 , v2 , v3 , v4 } where
v1 = (3, −6, 3, −9)
v2 = (−4, 6, 0, 2)
v3 = (−2, 4, −2, 6)
v4 = (6, −9, 0, −3)

13
 Find dim W and a basis for W.

78. Find the rank and nullity of A, and a basis for row(A), col(A), and null(A);

1 −2 4 1
 
 2 −3 9 −1 
A =  
1 0 6 −5 
 

2 −5 7 5

Find rank and nullity of (AT ) and confirm the Dimension Theorem.

79. Below are the sizes and ranks of the matrix A. Fill in the dimensions of the indicated
subspaces.
size of A 2 × 4 3 × 2 5 × 7
rank(A) 1 2 3
row(A)
col(A)
null(A)
null(AT )

80. What is the dimension of the subspace W = span(S) where

S = {(1, −1, 1, 2), (2, −2, 0, 1), (1, −1, −1, −1), (−4, 2, 0, 1)}

81. Find S ⊥ and (S ⊥ )⊥ where

S = {(−2, 3, 1, 1, 0), (1, 4, 2, 1, 2), (6, 2, 2, 0, 4)}

(To find S ⊥ , use the relation row(A)⊥ = null(A) where A is a certain matrix. To find
(S ⊥ )⊥ , find a basis for it.)

82. Extend the set S = {u1 , u2 , u3 } to a basis of R4 where

u2 = (1, 1, 3, 4)
u2 = (2, 7, 8, −4)
u3 = (2, 3, 5, −1)

83. Find the orthogonal projection of the vector v = [1 2 3]T along the vector a = [2 −3 2]T .
v·a
Do this in two ways; (i) use the formula proja v = a, and (ii) find the standard
kak2
matrix [T ] for this projection and compute T (v).

14
84. Show that v ∈ W ⊥ if and only if v · wi = 0, i = 1, . . . , k where W = span{w1 , . . . , wk }.
85. Let W = span{u1 , u2 }, where

u1 = (2, 2, 2), u2 = (1, −1, 1)

and v = (1, 0, 2). Write v = v1 + v2 where v1 ∈ W, v2 ∈ W ⊥ . Verify that v1 ∈ W by

writing it as a linear combination of u1 , u2 , and verify that v2 ∈ W ⊥ .
86. Find the orthogonal projection of v = (1, 1, 1, 1) onto null(A) where
3 −1 1 2
 
 −2 1 3 1 
A = 

−1 1 7 4
 
 
−5 2 2 −1
Verify that v − projnull(A) v is in row(A).

87. Find the closest point in the plane 3x − y + z = 0 to the point p = (1, 1, 1). What is
the distance from p to this plane?
88. Find the closest point on the line x(t) = (1, 2, 3) + t(−1, 0, 1) to p = (1, 1, 1). (You have
to shift the line so that it goes through the origin first..... and then shift the point p
too.)
89. Find the ”best” (i.e., least squares) solution to the system;
x1 − 8x2 − 7x3 − 4x4 = 2
2x1 − 3x2 − x3 + 5x2 = 3
3x1 + 2x2 + 5x3 + 14x4 = 0

Find the ”best” solution to the system;

x1 − 8x2 − 7x3 − 4x4 = 2
2x1 − 3x2 − x3 + 5x2 = 3
3x1 + 2x2 + 5x3 + 14x4 = 4

Which one of these systems is consistent? (check the error, i.e., kb̂ − bk)
90. Find the least squares line through the data;
(1, 1), (3, 2), (−2, 1), (4, 4), (5, 2).
Find the ’best’ quadratic through these points.
Make a sketch of your two curves and the data.
91. Use the Gram-Schmidt process to find an orthonormal basis for R3 beginning with the
basis {(1, 2, 3), (0, 1, 0), (−1, 0, 1)}
92. What would be the result if the Gram-Schmidt process was applied to a set of vectors
that was not linearly independent?
93. Let B1 = {(1, −1, 1), (2, 0, 1), (−2, 1, 1)} be a basis for R3 . Let v = (2, 3, 1).
Find (v)B1 (coordinates of v with respect to basis B).
If (w)B = (2, 3, 1), what is the vector w? (i.e., its coordinates in the standard basis).

15
94. (a) Let T : R3 → R3 be defined by

w1 = 2x1 + 3x2
w2 = −3x1 + x2 + 2x3
w3 = 4x1 − 3x2 − 2x3

Find [T ]B1 where B1 is given above.

Confirm your answer by verifying the formula [T (v)]B1 = [T ]B1 [v]B1 for v given above
(i.e. compute the right side and then compute the left side to see that they agree).
(b) Let B2 = {(3, 2, 0), (−2, 1, 1), (1, 0, 1)} be another basis of R3 . Find the change of
basis matrix PB1 →B2 . Use it to compute (v)B2 where v is given above, via the formula
[v]B2 = PB1 →B2 [v]B1 . Check your answer by computing [v]B2 directly.
(c) Verify the formlua

[T ]B2 = P [T ]B1 P −1

where P = PB1 →B2 . That is, compute [T ]B2 directly and compare the answer to the
result you get from the formula (same T as above).
(d) Now verify that

[T (v)]B2 = [T ]B2 [v]B2 = PB1 →B2 [T ]B1 [v]B1 = PB1 →B2 [T (v)]B1

95. (a) The two matrices below are similar. Show that they have the same determinant,
same eigenvalues, same rank, and same nullity. However, show that they do not have
the same eigenspaces.
   
0 −1 0 2 −27/4 9/2
A= 0

0 1 
, C =  −8/3

9 −8 
−4 −17 8 −4/3 11/2 −3

(b) Now show that A is diagonalizable (same A as above) and find the diagonal matrix
D it is similar to. Also find the matrix P that diagonalizes A (i.e., D = P −1 AP ). Is
the matrix C (as above) diagonalizable? If so, find the diagonal matrix H it is similar
to and the matrix M that diagonalizes it.

16
 Hints and answers

1. (1)T,T,F,F,(5)F,F,F,F,F,(10)F,F,T,T,F,(15)F,T,F,T,F,(20)F,F,T,T,T,(25)T,F,T,T,F,(30)T,F,T
2. (20, −10, 9, −15, −14); no
(31, −15, 25, 3, 3); yes
(30, −16, 28, 3, 6); yes
3.

x1 = 1 − 2t1 − t2
x2 = t1
x3 = 2t2 − t3
x4 = −3 − t2
x5 = t2
x6 = t3
t1 , t2 , t3 ∈ R

Example: t1 = 1, t2 = 2, t3 = 3; (−3, 1, 1, −5, 2, 3)

1 1
4. The system has the unique solution (1, c+3 , c+3 ) when c 6= −3 or 2, no solution when
c = −3, and many solutions (0, 1, 0) + t(5, −4, 1), t ∈ R when c = 2.
5. If k 6= 2 and for any p, there is a unique solution x = k−p
k−2
p−2
, y = k−2 , for k = 2 and p = 2
there are many solutions x = 1 − t, y = t, t ∈ R, and for k = 2 and p 6= 2 there are
no solutions.
6. x = (2, 3) + t(−1, 2), or x = (1, 5) + t(1, −2), or x = (3, 1) + t(−2, 4).
Some points on the line; (−1, 9) and (12, −17).
(1, −2) is not on this line. (−2, 11) is on the line.
7. x = (0, −3) + t(1, 2)
8. y = −x − 1
9. x = (1, 2, 3) + t(3, 1, 2). Some points; (4, 3, 5), (−8, −1, −3).
(2, 0, 1); this point is not on the line.
10. Intersection point is (5, 3).
11. (a)

2t − s = −1
4t + 2s = −1

(b)

2t + s = −1
4t + 2s = −1
4t + 3s = −2

17
12.
a11 t + a12 s = b1
a21 t + a22 s = b2
a31 t + a32 s = b3

Here is a line that will not intersect this plane;

(1, 1, 1) + t(1, −1, 0)

13. Since there are two variables (t and s), if there are 2 equations (2 lines) there can be
a unique solution (and hence an intersection), however, if there are more than 2 rows
(more than 2 lines) the system is most likely inconsistent. So, 2 lines typically intersect
in 2 dimensions, but not in 3 or higher dimensions.
14. NO, YES, NO
15. The equation of the plane is x + y + 2z = 9.
Some points on the plane; (0, 0, 4.5), (0, 3, 3). (2, −2, 3) is not on the plane.
Point of intersection is (−8, 5, 6).
To determine the angle of intersection between a plane P and a line L, imagine looking
at the plane edge on. The angle of intersection is the smallest angle between the plane
and the line, denoted by θ in the figures below. To determine this, consider the angle
between the two vectors n, a normal to the plane, and v, the direction of the line.

Note that there are two possibilities for the orientation of the normal of the plane n
and the
 direction
 vector v of the line. One case (left figure) is when the angle φ =
−1 n·v
cos knkkvk
between them is less than 90o (n · v > 0). Here the angle θ between the
 
line and the plane is θ = 90o − cos−1 knkkvk
n·v
. The second case (right figure) is when the
angle φ between the direction vectorof the line and the normal of the plane is greater
than 90o (n · v < 0). Here θ = cos−1 knkkvk
n·v
− 90o .

16. x = ( 10
17
, − 16
17
, 0) + t(11, 13, 17)
17. The unique solution is (x, y, z) = (−68, −82, −106)
18. The system of equations to determine the intersection of 4 planes would contain 4
equations in 3 variables. We usually expect such a system to be inconsistent, and
therefore, we would usually expect no intersection between 4 planes. (Recall from the
previous problem that usually 3 planes would intersect in a point; the fourth plane would
have to pass through this point in order for the 4 planes to have an intersection.)
19. x + y − z = 0
20. If a 6= 5, the point of intersection is
4a + 2 8 − 2a 3a2 − 2a + 5
x= , y= , z=
a+5 5+a 5+a

18
21. (a) The equations of the planes are
(t2 − t1 )x + (t1 + t2 )y + t1 z = t2 , t1 , t2 ∈ R

(b) Equations of the plane are t(x + y) = t, t ∈ R, or just x + y = 1.

22. (i)
1 − 81
   
1 1 1 −1 1 0
 2

0 2 −1 
 → 

0 1 0 − 12 

3
−3 2 1 −1 0 0 1 −8
Equation of the plane; x + 4y + 3z = 8.
23. (a) w = − √414 (−1, −2, 3)
√
(b) k = 20 ± 385 ≈ 39.62, 0.38.
24.
x1 = −10 + t
x2 = −12 + t
x3 = −2 + t
x4 = t

Yes, there is a positive solution; for t > 12.

25. p(x) = − 16 x3 + 16 x2 + x + 2
26. p(x) = ( 76 − 12 t)x2 + ( 11
6
− 12 t)x + t, t∈R
27. C3 H8 + 5 O2 = 3 CO2 + 4 H2 O
28. Decision variables; x1 , x2 , x3 , x4 denote the amounts of potatoes, rice, corn, and beans
respectively, in units of 100 grams.
Constraints;
carbohydrates : 12x1 + 8x2 + 5x3 + 15x4 = 40
protein : 4x1 + 6x2 + 5x3 + 15x4 = 25
fat : 4x1 + 2x2 + 5x3 + 6x4 = 15
System is
9 5
   
12 8 5 15 40 1 0 0 − 16 4
9 5 
 4 6  → 
5 15 25  0 1 0
 
4 2 
3
4 2 5 6 15 0 0 1 4
1

5 9
x1 = + t
4 16
5 9
x2 = − t
2 4
3
x3 = 1− t
4
x4 = t

19
 10
Since we are interested only in positive solutions, we see that 0 ≤ t ≤ 9
.
76 23
Note that the amount of the mixture is x1 +x2 +x3 +x4 = 16
− 16 t. So only this amount,
when 0 ≤ t ≤ 109
, can be produced.

29. If A is any 4 × m matrix, then in the product CA, the second row is zero, so this product
could never be the identity matrix.

30. CB −1 A, B −1 C T
 
31. H = − B + 12 B(C −1 )T (AT )−1
 
10 4 6
1
32. Y = 8  22 12 −6 

12 4 −2
33. subtract 2× row 1 from row 2, interchange rows 3 and 4, multiply row 2 by -1, add −5×
row 2 to row 3
 
0 −1 0
34. C = 

1 0 0 

0 −1 1
 
1 0 0 0 0
0 0 0 1 0
 
 
 
35. E = 
 0 0 7 0 0 


 0 1 0 0 0 

−4 0 0 0 1
 
1 0 0 0 0
0 0 1 0 0 
 

 
F = 
 0 1 0 0 0 


 0 0 0 1 −2 

0 0 0 0 1
36.
   
1 0 0 1 0 0
 −2
E1 =  1 0 
 E1−1 = 
 2 1 0 

0 0 1 0 0 1
   
1 0 0 1 0 0
E2 =  0 1 0  E2−1 =  0 1 0 
 
 
−4 0 1 4 0 1
   
1 0 0 1 0 0
E3 =  0

0 1 
 E3−1 = 
 0 0 1 

0 1 0 0 1 0
   
1 0 0 1 0 0
E4 =  0 1 0  E4−1 =  0 1 0 
 
 
0 1 1 0 −1 1

20
    
1 0 0 1 0 0
E5 =  0 1 0  E5−1 =  0 1 0 
 
 
0 0 −1 0 0 −1
   
1 0 −2 1 0 2
E6 =  0 1 0  E6−1 =  0 1 0 
 
 
0 0 1 0 0 1

A−1 = E6 E5 E4 E3 E2 E1
A = E1−1 E2−1 E3−1 E4−1 E5−1 E6−1

37. b1 = 2b2 − b3 ;

b = t1 v1 + t2 v2 , where v1 = (2, 1, 0), v2 = (−1, 0, 1)

Therefore, {v1 , v2 } is a basis for the range of A.

38. No. Yes. Yes.

39. x = t(3, 37 , 11
3
, 1)
x = (0, 0, 0, 1) + t(3, 37 , 11
3
, 1)

40. w = −6v1 + 3v2 + 2v3 . For the second, w is not a linear combination of v1 , v2 , v3

41. a − 2b − 32 c = 0
This gives us an expression for a basis for span{(2, 1, 0), (1, −1, 2), (0, 3, −4)};

{(2, 1, 0), (3, 0, 2)}

42. (1, −2, 5, −3), (2, 3, 1, −4), (0, 0, 0, 10) is a maximally linearly independent set.
The first three rows of B form a maximally linearly independent set.

43. The vector u4 is a linear combination of u1 , u2 , u3 ;

u4 = u1 + u2 − 2u3

44. (a) The set S is linearly dependent, and dim(span(S)) = 3.

(b) No.
(c) b = t1 (1, 12, 0, 0) + t2 (0, −2, 1, 0) + t3 (0, −1, 0, 1)

45. {(1, −3, 2), (2, −4, −1)} forms a basis for span{(1, −3, 2), (2, −4, −1), (1, −5, 7)}. The
dimension of this subspace is 2.

46.

47.

48.
19
49. x = ( 128 , − 63 , 79 , 5 )
64 96 16

21
50. -1

51. 82
1
52. 8 25

53. 6, -2, can’t determine

54.

55.

56. x3 = −80

57. A basis for a⊥ is {v1 , v2 , v3 , v4 } where

v1 = (1, 1, 0, 0, 0), v2 = (2, 0, 1, 0, 0), v3 = (4, 0, 0, 1, 0), v4 = (5, 0, 0, 0, 1).

58. Area = 5

59. Volume = 21

60. w = (−5, −11, −7)

√
61. 24 − 23i, −4 + 46i, −9 − 40i, −22 − 5i, − √6641 − √21 i,
41
533
√ −iθ   √
62. z = 13e , θ = tan−1 23 , or z = 13eiφ , φ = 2π − θ. Principle argument is −θ.
φ φ 2π φ 4π
z 9 = 139/2 e−9iθ . z 1/3 ; w1 = 131/6 ei 3 , w2 = 131/6 ei( 3 + 3 ) , w3 = 131/6 ei( 3 + 3 )
 √   √ 
3 23 3 23
63. 4
+i 4
, 4
−i 4

64. λ = 4, −2, −2. E4 = span{(− √16 , − √16 , √26 )}, E−2 = span{v1 , v2 }. λ = 4 has algebraic
multiplicity 1, geometric multiplicity 1. λ = −2 has algebraic multiplicity 2, geometric
multiplicity 2.

65. Bv1 = (−3, −3, −2) 6= λv1 , so v1 = (1, 2, 2) is not an eigenvector of B.

v2 = (2, 2, 0) is an eigenvector with eigenvalue λ = −2.
v3 = (1, −1, 1) is not an eigenvector. v4 = (0, 1, 1) is an eigenvector with λ = 4.
√ √
66. p(x) = 2(x + 1)(x + 2 − 5))(x + 2 + 5)

67. For both, they could be 1,2,3,4 or 5, but if the algebraic multiplicity of λ is k (k = 1, 2, 3, 4
or 5), the geometric multiplicity of λ must be ≤ k.

68. (a) Any vector w ∈ W is an eigenvector with eigenvalue 1. Any vector v ∈ W ⊥ is an

eigenvector with eigenvalue 0. det([T ]) is usually 0.
(b) If w is in this plane, then T (w) = w, and so w is an eigenvector with eigenvalue 1.
det([T ]) = 0.

69. F ; no. T ; no. T (x1 , x2 , x3 ); yes. G; yes.

70. Check whether AAT = I.

71. Only (c), (d) are orthogonal.

22
72. The first transformation clearly preserves the lengths of vectors, and hence is orthogonal.
Shears do no preserve the length of vectors.

73. (1) Not 1-1. Range is yz-plane (so not onto).

(2) Not 1-1 and not onto.
(3) Not 1-1 and not onto.

74.

75. First pair: T1 ◦ T2 ; yes. T2 ◦ T1 ; yes. Both define R3 → R3 .

Second pair: T1 ◦ T2 ; yes; R2 → R3 . T2 ◦ T1 ; no.
 √ 
3/2 0 1/2 " #
0 1 0
76. (a) Refer to page 12, Lecture 23; T1 =  0 1 √ 0 , T2 = . Note
 
1 0 0
−1/2 0 3/2
that T2 , geometrically, is projection down to the xy-plane and then reflection across the
diagonal y = x.
To find [T2 ◦ T2 ], refer to page 12 Lecture 23 and determine " (visually)#
0 1 0
T2 (T1 (1, 0, 0)), T2 (T1 (0, 1, 0)), T2 (T1 (0, 0, 1)) to find √
3/2 0 1/2
Now compare this to [T2 ][T1 ].

77. dim(W) = 2. Find two linearly independent vectors from {v1 , v2 , v3 , v4 } to form a basis
for W.

78. rank(A) = 2, nullity(A) = 2. rank(AT ) = 2, nullity(AT ) = 2.

size of A 2×4 3×2 5×7

rank(A) 1 2 3

79. row(A) 1 2 3
col(A) 1 2 3
null(A) 3 0 4
null(AT ) 1 1 2

80. basis = {(1, −1, 1, 2), (2, −2, 0, 1), (−4, 2, 0, 1)}. dim(W ) = 3.

81. S ⊥ = span{(−2, −5, 11, 0, 0), (1, −3, 0, 11, 0), (−6, −4, 0, 0, 11)}.
(S ⊥ )⊥ = span{(−4, 6, 2, 2, 0), (3, 1, 1, 0, 2)}.

82. {(1, 1, 3, 4), (2, 7, 8, −4), (2, 3, 5, −1), (41, 30, −33, 7)}
4 6 4
83. proja v = ( 17 , − 17 , 17 ).

84.

85. v = ( 32 , 0, 25 ) + (− 12 , 0, − 21 ).
1
86. projnull(A) v = 221
(88, 247, −25, 4).
9 3 3
87. Closet point is p̃ = ( 11 , − 11 , 11 ). Distance to plane is kp − p̃k.
√
88. Closest point is (2, 2, 2), distance = 3.

23
 q
89. x̂ = (− 1639
, − 11
39
, 0, 0) + t1 (−1, −1, 1, 0) + t2 (−4, −1, 0, 1). Error = 249
.
18 1
x̂ = ( 13 , − 13 , 0, 0) + t1 (−1, −1, 1, 0) + t2 (−4, −1, 0, 1). Error = 0; consistent system.

90. y = 1.36 + (0.29)x.

y = 1.31 + (0.26)x + (0.01)x2
√ √
91. {(0, 1, 0), √12 (−1, 0, 1), √ 1
√ (2+ 2, 0, 3 2−2)}. Here, we should begin using (0, 1, 0)
28−8 2
as the first vector, then (−1, 0, 1) and finally (1, 2, 3).

92. Proceed as in the algorithm. When at some point you encounter a vector that is in the
span of the previous ones (since a linearly dependent set has one or more vectors that
are in the span of the others), when you project that vector into the subspace of the
span of the previous ones, the resulting vector (residual) will be the zero vector. So you
just skip that vector (remove it from the basis) and go on to the next vector. In the end
you will produce an orthonormal set of vectors that has the same span as the original
set of vectors.

93. vB1 = (− 12 , 14 , 3 ). w = (6, −1, 6).

5 5 5
 
19 24 −61
1
94. (a) [T ]B1 =  −3 2 12 . [T (v)]B1 = (−3, 4, −4).

5
9 4 −16
 
−3 1 0
1
(b) PB1 →B2 =   −3 −2 9 .

[v]B2 = ( 10 , 7 , 2 ).
9 9 9
9
12 11 0
 √ 
2− 3 0√ 0
95. D =  0 2 + 3 0 ,
 


0 0 4
−1√ −1√
1
 7−4−1√
3 7+4 3
1√

P =  −2+ 3 2+ 3 4 


1 1 16

24