Contents

1. UNIT 1: VECTORS IN Rn AND Cn .........................................................................................................................3

1.1. Learning Outcomes ...................................................................................................................................3
1.2. Introduction ..............................................................................................................................................3
1.3. Vectors in Rn ..............................................................................................................................................5
1.4. Vector Addition and scalar Multiplication ................................................................................................5
1.5. Dot product ...............................................................................................................................................7
1.6. Norm and Distance in Rn ...........................................................................................................................7
1.7. Complex Numbers ....................................................................................................................................9
1.8. Vectors in Cn ........................................................................................................................................... 10
2. UNIT 2: LINEAR EQUATIONS .......................................................................................................................... 14
2.1. Learning outcomes ................................................................................................................................ 14
2.2. Introduction ........................................................................................................................................... 14
2.3. Linear Equations .................................................................................................................................... 14
2.4. System of Linear Equations.................................................................................................................... 15
2.5. Solution of a system of Linear Equations............................................................................................... 16
2.6. Solution of a Homogeneous System of Linear Equations. ..................................................................... 19
3. UNIT 3: VECTOR SPACES AND SUBSPACES..................................................................................................... 22
3.1. Learning Outcomes ................................................................................................................................ 22
3.2. Introduction ........................................................................................................................................... 22
3.3. Examples of Vector spaces .................................................................................................................... 23
3.4. Subspaces............................................................................................................................................... 24
3.5. Linear Combinations, Linear Spans ........................................................................................................ 26
3.6. Row Space of a Matrix ........................................................................................................................... 27
3.7. Sum and Direct Sums ............................................................................................................................. 28
4. UNIT 4: EIGENVALUES AND EIGENVECTORS .................................................................................................. 33
4.1. Learning Outcomes ................................................................................................................................ 33
4.2. Introduction ........................................................................................................................................... 33
4.3. Polynomials of matrices and linear operators ....................................................................................... 33
4.4. Eigenvalues and Eigenvectors ................................................................................................................ 34
4.5. Diagonalisation and Eigenvectors .......................................................................................................... 35
4.6. Characteristic Polynomial ...................................................................................................................... 36
4.7. 4.7 Minimum Polynomial ....................................................................................................................... 38
4.8. Characteristic and Minimum Polynomial of Linear Operators .............................................................. 39
5. REFERENCES ................................................................................................................................................... 42

2
1. UNIT 1: VECTORS IN Rn AND Cn
1.1. Learning Outcomes
Upon completing this unit you should be able to:
1. Solve systems of equations.
2. Compute dot products
3. Perform vector addition and scalar multiplication.
4. Find distances between vectors
5. Find norms of vectors
6. Simplify complex multiplication and division.

1.2. Introduction
In various physical applications there appear certain qualities such as temperature and
speed which possess only “magnitude.” These can be represented by real numbers and are
called “scalars”. On the other hand, there are also qualities, such as force and velocity,
which possess both “magnitude” and “direction”.
These quantities can be represented by arrows (having appropriate lengths and directions
and emanating from some reference point 0, and are called vectors. In this unit we study
the properties of such vectors in some detail.

We begin by considering the following operations on vectors.

(i) Addition: The resultant u + v of two vectors u and v is obtained by parallelogram
rule or law, i.e u + v is the diagonal of the parallelogram formed by u and v as
shown on the right.

v u+v

0 u

3
 i. Scalar multiplication:
The product ku of a real
-2u u 3u

number k by a vector u is obtained by multiplying the magnitude or length of u by a

vector u and retaining the same directions if k ≥ 0 or the opposite is k ˂ 0 as shown on
the right.
Now we assume the reader is familiar with the representation of the points in the plane by
ordered pairs of real numbers. If the origin of the axes is chosen at the reference point, 0
above, then every vector is uniquely determined by coordinates of its point. The
relationship between the above operations and end points follows.
i. Addition: if (a, b) and (c, d) are the endpoints of the vectors u and v, then (a + c,
b+d) will be the endpoint of u + v, as shown in Fig. (a) below.

(a + c, b+d) (ka,
kb)
ku
(c,d) u+v
v (a,b)
(a,b)
u
u

Fig. (a) Fig. (b)

ii. Scalar multiplication: If (a, b) is the end point of the vector u, then (Ka + Kb) will be
the end point t of the vector Ku, as shown in Fig. (b) above.
Mathematically, we identify a vector with its end point, that is, we call the ordered
pairs (a,b) of real numbers a vector. In fact we shall generalize this notion and call
an n- tuple

(a1, a2 ……………………an) of real numbers a vector. We shall again generalize the

coordinates of the n – tuple to be complex numbers and not just real numbers.
4
1.3. Vectors in Rn
The set of all n – tuples of real numbers, denoted by Rn, is called n –space A particular n –
tuple in Rn, say u = ( u1, u2, ………….un) is called a point or vector; the real numbers ui are
called the components (or coordinates) of the vector u. Moreover, when discussing the
space Rn we use the term “scalar’ for the elements of R, i.e. for the real numbers.

Example 1: Consider the following vectors:

(0,1), (1, -3), (1,2, √3, 4), (-5, ½, 𝜋).
The first two vectors have two components and so are points in R2; the last two vectors
have four components and so are points in R4.
Two vectors u and v are equal, written u = v if they have the same number of
components, i.e. belong to the same space, and if corresponding components are equal.
The vectors (1, 2, 3) and (2, 3, 1) are not equal since corresponding elements are not
equal.

Example 2: Suppose (x-y, x + z, z – 1) = (4, 2, 2). Then by definition of equality of

vectors,
x–y=4
x+y=2
z–1=3

1.4. Vector Addition and scalar Multiplication

Let u and v be vectors in Rn:
U = (u1, u2, …………..un) and v = (v1, v2, …………….vn)
The sum of u and v, written u + v, is the vector obtained by adding corresponding
components: u + v = (u1 + v1, u2 + v2……, un + vn).

The product of a real number K by the vector u, written ku, is the vector obtained by
multiplying each component of u by k:

5
 ku = (ku1, ku2, ……………….kun). Observe that u + v and ku are also vectors in Rn. We also
define,
-u = -1u and u – v = u + (-v).
The sum of vectors with different components is not defined.

Examples 3: Let u = (1, -3, 2, 4) and v – (4, 2, 1, 1).

Then u + v = ( 1+ 3, -3 + 5, 2 – 1, 4 – 2) = (4, 2, 1 , 2).

5u = (5.1, 5.(-3), 5.4) = (5,-15,10,20)

2u – 3v = (2,-6, 4, 8) + (-9, -15, 3, 6)= (-7,-21,7, 14)

Example 4 : The vector (0,0, ……, 0) in Rn, denoted by 0, is called the zero vector.
It is similar to the scalar O in that, for any vector u = (u1, u2………………, un),
u + o = (u1 + 0, u2 + 0, …………….un + 0)
= (u1, u2, …………,un) = u

Basic properties of the vectors in Rn under the operations of vector addition and scalar
multiplication are described in the following theorem.

Theorem 1.1 For any vectors u, v, w ∈ Rn and any scalars k, k ∈ R:

(i) ( u + v) + w = u + (v + w) (vi) ( k + k) u = ku + ku
(ii) u+o=u (vii) (kk) u = k(ku)
(iii) u + (-u) = 0 (viii) iu = u
(iv) u + v = v + u
(v) k(u + v) = ku + ku

Remarks: suppose u and v are vectors in Rn for which u = Kv for some non – zero scalar k ∈
r. Then u is said to have in the same direction as v if k ˃ 0, and in the opposite direction if k
˂ 0.

6
 1.5. Dot product
Let u and v be vectors in Rn:

u = (u1, u2, ………………………..un) and V = (v1, v2, ……………..vn).

The dot or inner product of u and v, denoted by u. v, is the scalar obtained by multiplying
corresponding components and adding the resulting products;
u.v = u1v1 + u2v2 + ……………unvn
The vectors u and v are said to be orthogonal(or perpendicular) if their dot product is zero:
u.v = 0

Example 5: Let u = (1, -2, 3, -4),

v = (6, 7,1, -2) and w = (5, -4, 5, 7).
Then u.v = 1.6 + (-2) .7 + 3.1 + (-4) .(-2) = 6 – 14 + 3 +8 =
u.w = 1.5 + (-2) . (-4) + 3.5 + (-4) .7 = 5 + 8 + 15 – 28 = 0
Thus u and v are orthogonal.
Basic properties of the dot product in Rn follow.
Theorem 12: For any vectors u, v, w ∈ Rn and any scalar k ∈ R:

i. (u + v) . w = u.w + v.w
ii. (ku) .v = k(u.v)
iii. u.v = v.u
iv. u.u ≥ 0 and u.u = 0 iff u = 0

Remark: The Space Rn with the above operations of vector addition, scalar multiplication and
dot product is usually called “Euclidian n – space.”

1.6. Norm and Distance in Rn

Let u and v be vectors in Rn: u = u1, u2, …………………………..un) and , v = (v1, v2, …..vn). The
distance between the points U and v, written d (u,v) is fined by

7
d(u,v) = √(𝑢1 – v1)2 + (u2 – v2)2 + ….+ (un – vn)2

The conjugate of the complex number z = (a, b) = a + bi is denoted and defined by z = a -

bi

Example 6; Let u = (1, -2, 4, 1) and v= (3,1,-5, 0).

Then d (u, v) = √(1 − 3)2 + (-2 -1)2 + (4 +5)2 +(1-0)2 = √95

|| v|| = √32 +12 + (-5)2 + 02 = √35

Now if we consider two points say P = (a, b) and q = (c, d) in the plane R2, then
|| P|| = √𝑎2 + b2 and d(p,q) = √(𝑎 − 𝑐)2 + (b – d)2 . That is, || P|| corresponds to the
usual Euclidian length of the arrow from the origin to the point p, and d (p, q) corresponds to
the usual Euclidian distance between the points p and q as shown below:

P=(a,b)
P=(a,b)
|b- d| √(𝑎-c)2+(b-d)2

√𝑎2+b2 |b| q = (c,d)

|a| |a-c|

A similar result holds for points on the line R and in space R3.

Remark: A vector e is called a unit vector if its norm is 1: ||e|| = 1. Observe that, for any non
– zero u∈Rn the vector en = u is a unit vector in the same direction as u.
||u||

Theorem 1.3 (Cauchy – Schwarz): For any vectors u, v ∈Rn, |u.v|≤||u|| ||v||.

8
Using the above inequality, we can now define the angle 𝜃 between any two non – zero
vectors
v ∈Rn by cos 𝜃 = u.v
||u| ||u||

Note that if u.v = 0, then 𝜃 = 90o or 𝜃 = 𝜋

2

1.7. Complex Numbers

The set of complex numbers is denoted by c formally, a complex number is an ordered pair
(a,b) of real numbers; equality addition and multiplication of complex numbers are defined
as follows:
(a,b) = (c,d) iff a = c and b = d
(a,b) + (c,d) = (a,c, b+d)
(a,b)(c,d) = (ac – bd, ad + bc)

The complex number (0,1) denoted

by i, has the important property that i2 = ii = (0,1) (0,1) = (-1,0) = -1 or i = √−1
Furthermore, using the facts,
(a,b) = (a,0) + (0,b) and (0,b) = (b, 0) (0,1),
We have (a,b) = (a,o) + (b,0) (0,1) = a+bi

The notation a + bi is more convenient than (a,b), for example, the sum and product of
complex numbers can be obtained by simply using the commutative and distributive laws
and i2 = -1:
(a + bi) (c +di) = a + c + c + bi + di = (a + c) + (b +d)i
(a + bi) (c + di) = ac + bci + adi + adi + bdi2

Example 7: Suppose z = (2 + 3i) + (5 – 2i) and w = 5 – 2i. Then

i. z + w = (2 + 3i) + (5 – 2i) = 2 + 5 + 3i – 2i = 7 + i

ii. zw = (2 + 3i) (5 -2i) = 10 + 15i – 4i – 6i2 = 16 + 11i

9
 iii. z = 2 + 3i = 2 – 3i and w = 5 – 2i = 5 + 2i

iv. w = 5 – 2i = (5 – 2i) (2 – 3i) = 4 – 19i

z 2 + 3i (2 + 3i) ( 2- 3i) 13 13

Just as the real numbers can be represented by the point on a line the complex numbers can
be represented by the points on the plane. Specifically, we let the point (a, b) in the plane
represent the complex number z = a + bi, i.e. whose real part is a and whose imaginary part
is b. The absolute value of z, written |z| is defined as the distance from z to the origin;
|z| = √(𝑎2 + b2

Example 8: Suppose z = 2 + 3i and w = 12 – 5i.

Then |z| = √4 + 9 = √13 and |w| = √144 + 25 = 1

1.8. Vectors in Cn
The set of all n – tuples of complex numbers, denoted by Cn, is called complex n – tuples.
Just as in the real case, the elements of Cn are called points or vectors; the elements of C are
called “scalars” and vector addition in Cn and scalar multiplication on Cn are given by;

(z, z2, ……..zn) + (w1, w2, ….., wn)

= (z1 + w1, z2 + w2, ……..zn + wn)
= (z1, z2, …………….zn) = (zz1, zz2, …..zzn)

Where zi, wi, z ∈ C

Example 9: ( 2 + 3i, 4 – i, 3) + (3 – 2i, 5i, 4 – 6i)

= (5 + i, 4 + 4i, 7 – 6i)
2i (2 + 3i, 4 – i, 3) = (-6 + 4i, 2 + 8i, 6i)

Now let u and v be arbitrary vectors in Cn:

10
u = (z1, z2, ….., zn), v = (w1, w2, ……………..wn) z; w;∈ c
The dot or inner product of u and v is defined as follows:
u.v = z1, w1, + z2w2 +……………+ znwn

Example 10; Let u = (2 + 3i, 4 – I, 2i) and

v= (3 – 2i, 5, 4 – 6i). Then

i. u.v = ( 2 + 3i) (3 – 2i) + (4 – i) (5) + (2i) ( 4 – 6i)

= (2 + 3i) (3 + 2i) + (4 – i) (5) + (2i) (4 + 6i)
= 13i + 20 – 5i – 12 + 8i = 8 + 16i

ii. u.u = (2 = 3i) (2 + 3i) + (4 – i) (4 –i) + (2i) (2i)

= (2 = 3i) (2 – 3i) + (4 – i) (4+i) + (2i) (-2i)
= 13 + 17 + 4 = 34

iii. || u|| = √𝑢. 𝑢 = 34

Exercise 1
1. Compute:
i. (3, -4, 5) + (1,1, -2)
ii. (1,2, -3) + (4, -5)
iii. -3 (4, -5, -6)

2. Let u = (2, -7, 1) , v= (-3, 0, 4)

W = (0, 5, -8). Find:
i. 3u – 4v
ii. 2u + 3v – 5w)
3. Find x and y if ( 4,y) = x(2,3)
4. Find x and y if (x,3) = (2,x,y)
5. Find x, y and z if (2, -3, 4) = x (1,1,1) + y(1,1,0) + z (1,0,0)

11
 6. Compute u.v where (i) u =(2,-3,6), v= (8,2,-3) (ii) u = (1,-8,0, 5), v = (3,6,4)
7. Determine k so that the vectors u and v are orthogonal where
i. u = (1,k,-3) and v=(2,-5,4)
ii. u = (2,3k, -4,1,5) and v= (6,-1,3,7,2k)
8. Find the distance d (u, v) between the vectors U and v where:
i. u = (1,7), v = (6, -5)
ii. u = (3, -5,4), v = (6,2,-1)
iii. u = (5,3,-2,-4,-1) , v = (2,-1,0,-7,2)
9. Find the norm ||u|| of the vector U if
i. u = (2,-7)
ii. u = (3,-12, -4)

Exercise 2
1. Simplify
i. (5 + 3i) (2 – 7i)
ii. (4 – 3i)2
iii. 1
3 – 4i

iv. 2 – 7i
5 + 3i
v. ( 1 + 2i)3
vi. 1
2 – 3i

2. Let z = 2 – 3i and w = 4 + 5i. Find:

i. z + w and zw
ii. z
w

iii. z and w
iv. | z| and |w|

12
3. Let u = (3 – 2i, 4i, H6i) and v = ( 5 + I, 2 – 3i, 5)
Find:
i. u+v
ii. 4iu
iii. (1 + i)v
iv. (1 – 2i)u + (3 + i)v
4. Find u.v and v.u where:
i. u = (1 – 2i, 3 + i), v = (4 + 2i, 5 – 6i)
ii. u = (3 – 2y, 4i, H6i), v = (5 + I, 2 – 3i, 7 + 2i)

13
2. UNIT 2: LINEAR EQUATIONS
2.1. Learning outcomes
Upon completing this unit you should be able to:
1) Reduce systems of equations to echelon form.
2) Solve systems of linear equations.
3) Solve homogeneous systems of linear equations.

2.2. Introduction
The theory of linear equations plays an important and motivating role in the subject of linear
algebra. In fact, many problems in linear algebra are equivalent to studying systems of linear
equations, e.g. finding the kernel of a linear mapping and characterizing the subspace
spanned by a set of vectors. Thus the techniques introduced in this unit will be applicable to
the more abstract treatment given later. On the other hand, some of the results of the
abstract treatment will give new insights into the structure of “concrete” systems of linear
equations.

2.3. Linear Equations

By a linear equation over the real field, it means an expression of the form
a1x1+a2x2+…+an xn = b
where the ai, b 𝜖 R and the xi are indeterminants (of unknowns of). The scalars ai are called
the “coefficients” of the xi respectively, and being called the constant term or simply constant
of the equation. A set of values ……. the unknowns, say
x1 = k1, x2 = k2, …xn = kn
is a solution of (1) if the statement obtained by substituting ki for xi,
a1k1+a2k2+ …an kn = b
is true. This set of values is then said to satisfy the equation. If there is no ambiguity about
the position of the unknowns in the equation, then we denote this solution by simply the n-
tuple
u = (k1, k2,…,kn).
EXAMPLE1: Consider the equation x+ 2y- 4z-k = 3
The 4 – tuple u = ( 3, 2, 1, 0 ) is a solution of the equation since
3 +2 × 2- 4 × 1 + 0 = 3 or 3 = 3 is a true statement. However, the 4 –tuple

14
V = (1, 2, 4, 5) is not a solution of the equation since 1+2×2-4×4+5=3 or -6 = 3 is not a
true statement.
Solutions of the equation (1) can be easily described and obtained. There are three cases:
Case (1): One of the coefficient in (1) is not zero say a ≠ 0. Then we can write the
equation as follows

a, x, = b – a2x2- …. –anxn. By arbitrarily assigning values to the unknowns x2…., xn, we

obtain a value for x1; these values form a solution of the equation. Furthermore, every
solution of the equation can be obtained in this way. Note in particular that the linear
equation in one unknown, ax + b, with, a ≠ 0 has the unique solution x = a-1b
Example 2: Consider the equation 2x – 4y + z = 8
We write the equation as 2x = 8 + 4y – z or x = 4 + 2y – 1 z
2
Any value of y and z will yield a value for x, and the three values will be a solution of the
equation. For example, let y = 3, z = 2; then x = 4 + 2 x 3 – 1 x 2 = 9
2
Case (ii): All the coefficients in (1) are zero, but the constant is not zero. That is, the
solution is of the form.
ox1 + ox2 + ……………….oxn = b, with b≠ o.
Then the equation has no solution.
Case (iii): All the coefficient in (1) are zero, and the constant is also zero. That is, the
equation is of the form.
ox1 + ox2 + …………………………….+oxn = 0
Then every n – turple of scalars in R is a solution of the equation.

2.4. System of Linear Equations

We now consider a system of m linear equations in the unknown x1, …….., xn:
a1x1 + a12x2 + …......a1nxn = b1
a21x2+a22x2 + …………………..a2nxn = b2
…………………………………………………………………

am1x1 + am2x2 +………amnxn = bm

Where the aij, bi belong to the real field R. The system is said to be “homogeneous” if the
constants b1, ………….bm are all zero. An n- turple u =(k1, ….., kn) of real numbers is a
solution set or the general solution.

The system of linear equations

15
a11x1 + a12x2 + …………………+ a1nxn = 0
a21x1 + a22x2 + ………………..a2nxn = 0
………………………………………………………………

am1x1 + am2x2 + …………………+ amnxn = 0

is called the homogeneous system associated with the fundamental relationship between
the systems. The above system always has a solution, namely the zero n- turple 0 = (0,
0,…….,0) called “the zero or trivial solution.” Any other solution, if it exists, is called a “non
zero or nontrivial” solution.

Theorem 1: Suppose u is a particular solution of the homogeneous system and suppose w

is the general solution of the associated homogeneous system.
Then u + w = { u + w: w 𝜖 W} is the general solution of the homogeneous systems.

2.5. Solution of a system of Linear Equations

Consider the above system of linear equations. We reduce it to a simpler as follows:
Step 1.Interchange equations so that the first unknown x1, has a non zero coefficient in the
first equation, that is so that a11 ≠ 0.
Step 2. For i > i, apply the operation
Li -a1L1 + a11Li
That is, replace ith linear equation Li, by the equation obtained by multiplying the first
equation Li by a11, and then adding.

Example 3: Consider the system of linear equations:

2x + 4y – z + 2v + 2w = 1
3x + 3y + z – v + 4w = -7
4x + 8y + z + 5v – w = 3
We eliminate the unknown x from the second and third equations by applying the following
operations:
L2 -3L1+ 2L2 and L3 -2L1 + L3
We compute -3L1: - 6x – 12y + 3z – 6v – 6w = -3

16
 2L2; 6x +12y + 2z – 2v + 8w = -14

-3L1 + 2L2: 5z – 8v + 2w = -17

and
-2Li: -4x – 8y + 2z – 4v – 4w = -2
L3: 4x + 8y + z + 5v – w = 3

-2LrL3: 3z + v – 5w = 1
Thus the original system has been reduced to the following equivalent system:
2x + 4y – z + 2v + 2w = 1
5z – 8v + 2w = -17
3z + v – 5w =1

Note that the above equations, excluding the first, form a system which has fewer
unknowns than the original system. Note that:
(i) If an equation ox1 + ….. +oxn = b, b ≠ 0 occurs, then the system is inconsistent
and has no solution:
(ii) If an equation ox1 + …oxn = 0 occurs, then the equation can be deleted without
affecting the solution.

Example 4: Reduce the following system by applying the operations L2 -3L1 + 2L2
and L3 -3L1 + 2L2, and then the operation L3 -3L2 + L3.

2x + y – 2z + 3w = 1 2x + y – 2z + 3w = 1
3x + 2y – z + 2w = 4 y + 4z – 5w = 5
28 + 3y + 3z – 3w = 5 3y + 12z – 15w = 7

2x + y – 2z = 3w = 1
Y + 4z – 5w = 5
0 = -8
The equation 0 = -8, that is 0x + 0y + 0z + qw = -8.
Shows that the original system is inconsistent and so has no solution.

Example 5: Reduce the following system by applying the operations

L2 -L1 + L2,

L3 -2L1 + L3 and L4 - - 2L1 + L4, and then the operations L3 L2 – L3 and

L4 -2L2 + L4:

17
x + 2y – 3z = 4 x + 2y – 3z = 4 x + 2y – 3z = 4
x + 3y + z = 11 y + 4z = 7 y +4z = 7
2x + 5y – 4z =13 y + 2z =5 2z = 2
2x + 6y + 2z = 22 2y + 8z = 14

We observe that the system is consistent since there is no equation of the form 0 = 16,
with b ≠ 0. Furthermore, since in echelon form there are three equations in the three
unknowns, the system has a unique solution. By the third equation z = 1. Substituting z =
1 into the second equation, we obtain y = 3. Substituting z = 1 and y = 3 into the first
equation . We find x = 1. Thus x = 1, y = 3 ans z = 1, or in other words, the 3 – tuple (1,
3, 1), is the unique solution of the system.

Example 6: Reduce the following system by applying the operations L2 2L1 + L2 and
L3 -5L1 + L3, and then the operation L3 -2L2 + L3:

x + 2y – 2z + 3w = 2 x + 2y – 2z + 3w = 2
2x + 4y – 3z + 4w = 5 z – 2w = 1
5x + 10y – 8z + 11w = 12 2z – 4w= 2

x + 2y – 2z + 3w = 2 x + 2y – 2z + 3w = 2
z – 2w = 1 z – 2w = 1
0=0

The system is consistent, and since there are more unknowns than equations in echelon
form, the system has an infinite number of solutions. In fact, there are two free variables,
y and w , and so a particular solution can be obtained by giving y and w any value. For
example, let w = 1 and y = -2.

18
Substituting w = 1 into the second equation, we obtain z = 3. Putting w = 1, z = 3 and y =
-2 into the first equation, we find x = 9. Thus x = 9, y = -2, z = 3 and w = 1, or in other
words, the 4 tuple (9, -2, 3, 1) is a particular solution of the system.

2.6. Solution of a Homogeneous System of Linear Equations.

If we begin with a homogeneous system of linear equations, the system is clearly
consistent since, for example it has the zero solution 0= (0,0,…………., 0). Thus it can
always reduce to an equivalent homogeneous system in echelon form:

a11x1 + a12x2 + …. a1nxn = 0

a2j2xi2 + a2j2+ ……………. + a2nxn = 0
…………………………………………………………..
arjxjr + ………………….. + arnxn = 0

Exercise 1
Solve the following Linear equations:
1. 2x – 3y + 6z + 2v – 5w = 3
y – 4z + v = 1
v – 3w = 2

2. x + 2y – 3z = -1
3x – y + 2z = 7
5x + 3y – 4y = 2

3. 2x + y – 2z = 10
3x + 2y + 2z = 1
5x + 4y + 3z = 4

4. x + 2y – 3z = 6
2x + 3y + az = 3
x + ay + 3z = 2

5. x – 3y + 4z – 2w = 5
2y + 5z + w = 2
y – 3z = 4
19
6. x + y – z = 1
2x + 3y + az = 3
x+ ay + 3z = 2

7. 2x + y – 3z = 5
3x – 2y + 2z = 5
5x – 3y – 3z = 16

8. 2x + 3y – 2z = 5
x – 2y + 3z = 2
4x – y + 4z = 1

9. x + 2y – z + 3w = 3
2x + 4y + 4z = 3w = 9
3x + 6y – z + 8w = 10

10. Determine the values of k such that the system in unknowns x, y and z has: (i) a
unique solution, (ii) no solution,
(iii) More than one solution:
(a) kx + y + z = 1 (b). x + 2y + kz = 1
x + ky + z = 1 2x + ky + 8z = 3
x + y + kz = 1

Exercise 2
1. Determine whether each homogeneous has a non zero solution:
a. X + 2y – 2z = 0 b. x + 3y – 2z = 0
X – 8y + 8z = 0 2x - 3y + z = 0
3x – 2y + 4z = 0 3x – 2y + 2z = 0

C. x + 2y – 5z + 4w = 0
2x – 3y + 2z + 3w = 0
4x – 7y + z – 6w = 0

2. Determine whether each homogeneous system has a non zero solution:

a. x – 2y + 2z = 0
2x + y – 2z = 0
3x + 4y – 6z = 0
3x – 11y + 12z = 0
20
b. 2x – 4y + 7z + 4v – 5w = 0
9x + 3y + 2z – 7v + w = 0
5x + 2y – 3z + v + 3w = 0
6x – 5y + 4z – 3v – 2w = 0

21
3. UNIT 3: VECTOR SPACES AND SUBSPACES
3.1. Learning Outcomes
Upon completing this unit should be able to:
1. Define a vector space or subspace.
2. Write vectors as linear combinations.
3. Show that v is or is not a vector space.
4. Determine subspaces.
5. Show that given vectors generate Rn.
6. Determine whether given matrices have the same column spaces.
7. Determine whether given matrices have the same row spaces.

3.2. Introduction
In unit 1 we studied the concrete structures Rn and Cn and derived various properties. Now
certain of these properties will play the role of axioms as we define abstract “Vector
Spaces”, or as they are sometimes called “Linear Spaces.”
The definition of a vector space involves an arbitrary field, K whose elements are called
“Scalars.” We adopt the following notation (Unless otherwise stated or implied).

K – the field of scalars

A, b, c,o,k – the elements of K
V – the given vector spaces
u, v, w – the elements of v.
Definition
Let k be a given field and let V be a non empty set with rules of addition and scalar
multiplication which assigns to any u, v ∈ V, a sum u + v ∈ V. Then V is called space
over k (and the elements of V are called vectors) if the following axioms hold:

A1: For any vectors u, v, w, ∈ V, ﴾ u +v ﴿+w ꞊ u +﴾ v+ w﴿.

22
 A2: There is a vector in V, denoted by 0 and called the zero vector, for which u + 0
= u for any vector u ∈ V.
A3: For each vector u ∈ V there is a vector in V, denoted by –u, for which u + (-u)
= 0.
A4: For any vector u, v ∈ V, u + v, = v + u.
M1: For any scalar k ∈ K and any vectors u, v ∈ V, k(u + v) = ku + kv.
M2: For any scalars a, b ∈ k and any vector u ∈ v, (a + b)u = au + bu.
M3: For any scalars a, b ∈ K and any vector u ∈ V, (ab)u = a(bu).
M4: For the unit scalar 1 ∈ K, 1u = u for any vector u ∈ V.
The above axioms naturally split in two sets. The first four are only concerned with the
additive structure of V and can be summarized by saying that V is commutative group and
under addition. It follows that any sum of vectors of the form
V1 + v2 + ……… + vm requires no parenthesis and does not depend upon the order of the
summads, the vector 0 is unique, the negative –u of u is unique, and the cancellation law
holds:
u + w = v + w implies u = v for any vectors u, v, w ∈ V. Also subtraction is defined by
u – v = u + (-v)
On the other hand, the remaining four axioms are concerned with the action” of the field K on
V.

3.3. Examples of Vector spaces

Now here is a list of important examples of vector spaces. The first example is a
generalization of the space Rn.

Example 1: Let K be an arbitrary field. The set of all n – tuples of elements of K with vectors
addition and scalar multiplication defined by
(a1, a2, …………….., an) + (b1, b2, …….bn) = (a, + b1, a2 + b2, ….., an + bn) and k(a1, a2,
…..an) = (ka1, ka2, ……,kan).

23
Example 2: Let V be the set all m x n matrices with entries from an arbitrary field K. then V
is a vector space over K with respect to the operations of matrix addition and scalar
multiplication.

Example 3: Let V be the set polynomials a0 + a1 + a2t2 + ……antn with coefficients ai from a
field K. Then V is a vector space over K with respect to the usual operations of addition of
polynomials and multiplication by a constant.

Example 4: Let K be an arbitrary field and let X be any non empty set. Consider the set V of
all functions from X into K. The sum of sum of any two functions f, g ∈ V is the function f + g
𝜖 V denoted by
(f + g (x) = f(x) +g (x)
And the product of a scalar k ∈ k and a function f ∈ v is the function kf ∈ v defined by
(kf) (x) = kf(x).

Then V with the above operations is a vector space over V. the zero vector in V is the zero
function 0 which maps each x 𝜖 X into 0 ∈ K: 0(x) = 0 for every x ∈ X. Furthermore, for any
function f ∈ V, -f is that function in V for which (-f) (x) = -f (x) for every x ∈ X.

Example 5: Suppose E is a field which contains a subfield K. Then E can be considered to be

a vector space over K, taking the usual addition in E to be the vector addition and defining
the scalar product kv of k ∈ E to be the field E. Thus the complex field C is a vector space
over the real field R, and the real field R is a vector space over the rational space Q.

3.4. Subspaces
Let W be a subspace of a vector space over a field K. W is called a subspace of V if W is itself
a vector space over K, with respect to the operations of vector addition and scalar
multiplication on V.
Simple criteria for identifying subspaces follow.
Theorem 2: W is a subspace of V if and only if

24
 i. W is a non empty,
ii. W is closed under vector addition: v, w ∈ W implies v + w ∈ W,
iii. W is closed under scalar multiplication v ∈ W implies kv ∈ W for every k ∈ K.

Example 6
Let V be any vector space. Then the set {0} consisting of those vectors whose third
component is zero, W = {a, b, o}: a, b ∈ R}, is a subspace of v.
Example 7:
(i) Let V be the vector space R³. Then the set W consisting of those vectors whose
third component is 0, W = (a,b,0):a,b ∈ 𝑅, is a sub space of V.
(ii) Let b be the space of all square n x n matrices. Then the set W consisting of
those entries A = (aij) for which aij = aji, called symmetric matrices, is a
subspace of V.
(iii) Let v be the space of polynomials. Then the set W consisting of polynomials with
degree ≤ n, for a fixed n, is a subspace of V.
iv. Let v be the pace of all functions from a non empty set X into the real field R. Then
the set W consisting of all bounded functions in V is a subspace of v. (A function f
∈V is bounded if there exists M ∈ R such that |f(x)| ≤ m for every x ∈ V).

Example 8. Let U and W be subspaces of a vector V. We show that the intersection U∩W is
also a subspace V. Clearly O∈U and O∈W since U and W are subspaces; where O∈U∩W. Now
suppose u, v∈U∩W. Then u, v∈U∩W and u, v∈W and since U and W are subspaces, Au +
bv ∈ v and au + bv ∈ W for any scalars a, b ∈ K. Accordingly au + bv ∈ U ∩ W and so U ∩ W
is a subspace of v.

The result in the preceding example generalizes as follows.

Theorem 4: The intersection of any number of subspaces of a vector space is a subspace of
V.

25
 3.5. Linear Combinations, Linear Spans
Let V be a vector space over a field K and let v 1, ……., vm ∈ V. Any vector in V of the form
a1v1 + a2v2 + ……+ amvm where the ai ∈K, is called a linear combination of v1, ….., vm. The
following theorem applies.
Theorem 5. Let S be a .00.0 non empty subset of V. The set of all linear combinations of
vectors in S, denoted by L(S) is a subspace of V containing S. Furthermore, if W is any
other sub space of V containing S, then L(S) ⊂W.
In other words, L(S) is the smallest subspace of V containing S, hence it is called the
subspace “Spanned or generated by S. For convenience (∅) is defined as {0} i.e L(∅) = {0}

Example 9. The vector e1 = (1,0,0), e2 = (0,1,0) and e3 = (0,0,1) generate the vector
space R3 for any vector (a,b,c) ∈R3 is a linear combination of the eij, specifically, (a,b,c) =
a(1,0,0) + b(0,1,0) + c(0,0,1)
= ae1 + be2 + ce3

Example 10: The polynomials 1, t, t2, t3…….generate the vector space V of all polynomials
(in t): V = L (1,t,t2,…..) for any polynomial is a combination of 1 and powers of t.

Example 11: Determine whether or not the vector v = (3, 9, -4, 2) is a linear combination
of the vector u1 = ( 1, -2, 0, 3), u2 = (2,3,0, -1) and u3 = (2, -1, 2,2). i.e. belongs to the
space spanned by the ui.

Set v as a linear combination of the ui using unknowns x, y and z; that is set v = xu1 + yu2
+ zu3:
(3, 9, -4, -2) = x (1, -2, 0, 3) + y (2, 3, 0, -1) + z (2, -1, 2,1)
= (x + 2y + 2z, -2x + 3y – z, 3x, -y + z)
Form the equivalent system of equations by setting corresponding components equal to
each other, and then reduce to echelon form:

x + 2y + 2z = 3 x + 2y + 2z = 3 x + 2y + 2z = 3
-2x + 3y – z = 9 or 7y + 3z = 15 or 7y + 3z = 15

26
 2z= -4 2z = -4 2z = -4
3x – y + z = -2 -2z = 4

or x + 2y + 2z = 3
7y + 3z = 15
2z = -4

Note that the above system is consistent and so has a solution; hence; v is a linear
combination of the ui. Solving for the unknowns we obtain x = 1, y = 3, z = -2.
Thus v = u1 + 3u2 – 2u3.
Note that if the system of linear equations were not consistent, i.e had no solution, then the
vector v would not be a linear combination of the ui.

3.6. Row Space of a Matrix

Let A be an arbitrary m x n matrix over a field K:

A = a11 a21 ….a1n

a21 a22 … a2n
am1 am2 …amn

The row of A,
R1 = (a11, a21, ….a1n), …..Rm = (am1, am2, …..,amn) viewed as vectors in Kn, span a subspace
of Kn called “the row space of A”. That is, row space A = L (R1,R2, …..Rm).
Analogously, the columns of A, viewed as vectors in Km, span a subspace of Km called the
column space of A.

Theorem 6: Row equivalent matrices have the same row space.

Theorem 7: Row reduced echelon matrices have the same row space if and only if they have
the same non zero rows.
Thus, every matrix is row equivalent to a unique row reduced echelon matrix called its “row
canonical form”.
We apply the above results in the next example.
27
Example 12: Show that the space U generated by the vectors
u1 = (1, 2, -1, 3), u2 = (2, 4, 1, -2) and u3 = (3, 6, 3, -7) and the space V generated by the
vectors v1 = (1, 2, -4,11) and v2 = (2, 4, -5, 14) are equal i.e U = V.
Method 1: Show that each Ui is a linear combination of v1 and v2, and show that each vi is a
linear combination of u1 and u2 and u3. Observe that we have to show that six stems
of linear equations are consistent.
Method 2: Form the matrix A whose rows are the ui and row reduce A to “row canonical
form:

A=
1 2 -1 3 1 2 -1 3 1 2 -1 3
2 4 1 2 to 0 0 3 -8 to 0 0 3 -8 to
3 6 3 7 0 0 6 -16 0 0 0 0

1 2 0 1
3
0 0 1 -8
3
0 0 0 0

Now form the matrix B whose rows are v1 and v2, and row reduce B to row canonical form:

1 2 -4 11 1 2 -4 11 1 2 0 1
B= 2 4 -5 14 to 0 0 3 -8 to 3
0 0 1 -8
3

Since the non zero rows of the reduced matrices are identical, the row spaces of A and B
are equal and so U = V.

3.7. Sum and Direct Sums

Let U and W be subspaces of a vector space V. The sum of U and W, written U + W

consists of all sums U + W where u ∈ U and w ∈ W:
U + W = { u + w: u ∈ W: u ∈ U, w ∈ W}
28
 Note that 0 = 0 + 0 ∈ U + W, since 0 ∈ U + W, since 0 ∈ U, 0 ∈ W, Furthermore, suppose
u + w and u’ + w’ belong to U + W, with u, u’ ∈ U and w, w’ ∈ W. Then (u + w) + ( u’ +
w’) = (u + u’) + (w + w’) ∈ U + W and for any scalar k, k(u + w) = ku + kw ∈ U + W.

Theorem 7: The sum U + W of the subspaces U and W of V is also a subspace of V.

Example 13: Let U consist of those matrices over R. Let U consist of those matrices in V
whose second row is zero, and let W consist of those matrices in V whose second column
is zero:

U= a b a 0
: a, b ∈ R , W = : a, c ∈ R
0 0 c 0

Now U and W are subspaces of spaces of V.

We have: U + W = a b : a, b, c ∈ R and U ∩ W = a 0 :a∈ R
C 0 0 0

That is, U + W consists of those matrices whose lower right entry is 0, and U ∩ W consists of
those matrices whose second row and second column are zero.

Definition: The vector space V is said to be the direct sum of its subspaces U and W,
denoted by V = U Φ W if every vector v ∈ V can be written in one and only one way as v = u
+ w, where u ∈ U and w ∈ W.

Theorem 8: The vector space V is the direct sum of its subspaces U and W if and onLy if:
(i) V = U + W, and (ii) U ∩ W = {0}.

Example 14: In the vector space R3, let U be the xy plane and W be the yz plane:

U = {(a,b,0): a,
b ∈ R} and W = {(0,b,c): b, c) : {(0,b,c,): b, C ∈ R}

29
Then R3 = U + W since every vector in R3 the sum of a vector in U and a vector in W.
However, R3 is not the direct sum of U and W since such sums are not unique; for example.
(5, 5, 7) = (3, 1, 0) + (0, 4, 7) and also (3, 5, 7) = (3, -4, 0) + (0, 9, 7)

Example 16: In R3, let U be the xy plane and let W be the z axis:
U = {(a, b, c),: a, b, ∈ R} and W = {(0, 0, 0,c): C ∈ R)}={(a,b,1):a, b, ∈ R}and w={(0,0,c):c
∈ R}
Now any vector (a,b,c), ∈ R can be written as the sum of a vector in U and a vector in V in
one and only way:
(a, b, c) = (a, c, 0) + (0, 0, c)
Accordingly, R3 is the direct sum of U and W, i.e. R3 = U Φ W

Exercise
1. Let V be the set of all functions from a non empty set X into a field K. for any functions
f, g ∈ v and any scalar. k ∈ K, let f + g and kf be the functions in v defined as follows:
(f + g) (x) = f(x) + g(x) and (kf)(x) = kf (x), vx ∈ X. ( The symbol ∀ means “for
every”). Prove that v is a vector space over k.
2. Let V be the set of ordered pairs of real numbers: V = {(a,b):a,b,∈ R}. show that 𝑉 is not
a vector space over R with respectively to each of the following operations of addition in V
and scalar multiplication on V:
(i) (a, b) + (c, d) = (a + c, b + d) and k(a ,b) = (ka, b)
(ii) (a, b) + (c, d) = (a ,b) and k(a ,b) = (ka, kb)
(iii) (a, b) + (c, d) = (a + c, b =d) and k(a, b) = (k2a, k2b).
3. Let V = R3. Show that W is not a subspace of V where;
(i) W = {(a, b, c):a ≥ 0)}, i.e W consists of those vectors whose first component
is non negative;
(ii) W = {(a, b, c):a2 + b2 + c2 ≤ 1}, i.e W consists of those vectors whose length
does not exceed 1.
(iii) W = {(a, b, c): a, b ∈ Q)} i.e. W consists of those vectors whose components
are rational numbers.

30
4. Let V be the vector space of a 2 by 2 matrice over the real field R. show that W is not a
subspace of V where:
(i) W consists of all matrices with zero determinant.
(ii) W consists of all matrices A for which A2 = A.
5. Write the vector v = (1, -2, 5) as linear combination of the vectors e1 (1, 1, 1), e2 =(1,2,3)
and e3 = (2, -1, 1).
6. Write the vector v = (2, -5, 3) in R3 as a linear combination of the vectors e1 = (1, -3m 2)
e2 = (2, -4,-1) and e3 = (1,-2, k) in R3 as a linear combination of the vectors e1 = (1, -
3,2), e2 = (2, -4, -1) and e3 = (1, -5, 7).
7. For which value of k will the vector u = (1, -2, k) in R3 be a linear combination of the
vectors v = (3, 0, -2) and w = (2, -1, -5)?
8. Write the polynomial v = t2 + 4t – 3 over R as a linear combination of the polynomials e1
= t2 – 2t + 5, e2 = 2t2 – 3t and e3 = t + 3.
9. Write the matrix E = 3 1 as a linear combination of the matrices A = 1 1 B = 0 0
1 -1 1 0 1 1
and C = 0 2
0 -1
10. Show that the vectors u = (1, 2, 3), v = (0, 1,2) and w = (0,0,1) generate R3.
11. Find conditions on a, b and c so that (a, b, c) ∈R3 belongs to the space generated by
u = (2, 1, 0), v = (1, -1, 2) and w = (0, 3, -4)
12. Show that the xy plane W = (a, b, 0) in R is generated by u and v where
i. u = (1, 2, 0) and v = (0, 1, 0)
ii. u = (2, -1, 0) and v = (1, 3, 0)
13. Determine whether the following matrices have the same row space

A= 1 1 5 B= 1 -1 -2 c = 1 -1 -1
2 3 13 , 3 -2 -3 , 4 -3 -1
3 -1 3

31
14. Determine whether the following matrices have the same column space:

A= 1 3 5 B= 1 2 3
1 4 3 , -2 -3 -4
1 1 9

15. Let U and W be subspaces of a vector space V. show that

(i) U and W are contained in U + W
(ii) U + W is the smallest subspace of V containing U and W, that is U + W is linear
space of U and W: U + W = L (U, W).
16. Prove that the vector space V is the direct sum of its subspaces U and W if and only if
(i) V = U + W and (ii) U ∩ W = {0}
17. Let U and W be the subspaces of R3 defined by U = {(a, b, c): a = b = c} and
W = {(0,b,c)}. Show that R3 = U Φ W.
18. Show that (1, 1, 1) and (0, 1, -1) generate R3, i.e. that any vector (a, b, c) is a linear
combination of the given vectors.

32
4. UNIT 4: EIGENVALUES AND EIGENVECTORS
4.1. Learning Outcomes
Upon completing this unit you should be able to:
1) Define the concepts Eigen values and eigenvectors.
2) Find Eigen values and eigenvectors
3) Find bases of Eigen spaces
4) Find characteristics of polynomials
5) Find minimum polynomials

4.2. Introduction
In this unit we investigate the theory of a single linear operation on a vector space V of
finite dimension. In particular, we find conditions under which T is diagonalizable.
We shall also associate certain polynomials with an operator T: its characteristic polynomial
and its minimum polynomials. These polynomials and their roots play a major role in the
investigation of T. The particular filed K also play an important part in the theory since the
existence of roots of a polynomial depends on K.

4.3. Polynomials of matrices and linear operators

Consider a polynomial f(t) over field K: f(t)=an tn+……….. a1 t + ao. If A is a square matrix
over K, then we define f(A)=anAn+…… a, A + aoI, where I is the identify matrix. In
particular we say that A is a root or zero of the polynomial f(t) is f(A)=0.
1 2
Example 1: Let A = ( ) and let f(t) =2t2 -3t +7,g(t) =t2 – 5t -2.Then
3 4
1 2 2 1 2 1 0 18 14
f(A) =2( ) -3( ) + 7( )=( ), and
3 4 3 4 0 1 21 39
1 2 2 1 2 1 0 1 0
g(A)= ( ) -5( )-2( )=( )
3 4 3 4 0 1 0 1
Thus A is a zero of g(t).
The following theorem applies.

33
 Theorem 1: Let f and g be polynomials over, and let A be an N-square matrix over
K. Then
i. ( 𝑓 + 𝑔)𝐴 = 𝑓 (𝐴) + 𝑔 (𝐴)
ii. (fg)(A)=f(A)g(A), and for any scalar k∈K,
iii. (kf) (A) =kf(A)

Furthermore, since f(t)g(t)=g(t) f(t) for any polynomials f(t) and g(t), f(t) g(A)=g(A)f(A).
That is any two polynomials in the matrix A commute.
Now suppose T : V V is a linear operator on a vector space V over K. If f(t) =𝛼𝑛𝑡n +…..+ a1,t
+ ao, then we define f(T) in the same way as we did for matrices: f(T)=anTn+…..+a1T+ aoI,
where I is now the identity mapping.
We also say that T is a zero or root of f(t) if f(T) = 0. The relations in theorem I hold for
operators as they do for matrices, hence any two polynomials in T commute. Furthermore if A is a
matrix representation of f(T), then f(A) is the matrix representation of f(T). In particular, f(T) =0,
if and only if f(A)=0.

4.4. Eigenvalues and Eigenvectors

Let: V V be a linear operator on a vector space V over a filed K. A scalar 𝜆 ∈K
is called an Eigen value of T if there exists nonzero vector v∈V for which T (v) =𝜆 𝑣.
Every vector satisfying this relation is then called an eigenvector of T belonging to the
eigenvalues 𝜆. Note that each scalar multiple KV is such an eigenvector:
T(KV) = KT(V)=K(𝜆 𝑣 ) = (KV) The set of all such vectors is subspace of V called the
eigen space of 𝜆 .
The terms characteristic value and characteristic vector (or proper value and
proper vector) are frequently used instead of eigenvalue and eigenvector.
Example 2:Let I : V V be the identify mapping. Then, for every v∈V, I)v) =v=Iv.
Hence 1 is an eigenvalue of I, and every vector in V is an eigenvector
belonging to I.
1 2
Example 3: Find eigenvalues and associated nonzero eigenvectors of the matrix A =( ).
3 2

34
 𝑥
We seek a scalar t and a nonzero vector X = 𝑦 such that AX = tx:

1 2 𝑥 𝑥
( ) 𝑦 =t 𝑦
3 2

The above matrix equation is equivalent to the homogenous system.

X +2y =tx ( t – 1) x -2y =0
3x +2y = ty or -3x + (t-2) y =0 (1)
Recall that the homogenous system has a nonzero solution if and only if the determinant of
the matrix of coefficients is 0:
t–1 -2 = t2 -3t – 4 = (t -4) (t+1) =0
-3 t-2

Thus t is an eigenvalues of A if and only if t = 4 or t =-1. Setting t = 4 in (1),

-2x - 2y =0 or simply 3x -2y =0
-3x +2y =0
𝑥 1
Thus w= 𝑦 = is a nonzero eigenvector belonging to the eigenvalue t =-1, and every
−1
eigenvector belonging to t= -1 is a multiple of w.
The next theorem gives an important characterization of eigenvalues which is frequently
used as its definition.
Theorem 2: Let T: V V be a linear operator on a sector space over K. Then 𝜆 ∈K is an
eigenvalue of T, if and only if the operator 𝜆 I, -T is singular. The eigenspace of
𝜆 𝑖𝑠 𝑡ℎ𝑒𝑛 𝑡ℎ𝑒 𝐾𝑒𝑟𝑛𝑒𝑙 𝑜𝑓 𝜆 𝐼 − 𝑇.
Theorem 3: Nonzero eigenvectors belonging to distinct eigenvalue are linearly independent.

4.5. Diagonalisation and Eigenvectors

Let T: V V be a linear operator on a vector space V with finite dimension n.
Note that T can be reprinted by a diagonal matrix:
K1 0 . . . 0
0 K2 . . . 0
…………………
0 0 Kn if and only if there exists a basis (v1,……….. vn) of V for which

35
 T (v1) = K1 v1
T (v2) = K2 v2
...................................
T(vn) = Kn vu.
That is such that the vectors v1,……, vn are eigenvectors of T belonging respectively to
eigenvalues K,… Kn. In other words:
Theorem 4: A linear operator T: V V can be represented by a diagonal matrix B if
and only if V has a basis consisting of eigenvectors of T. In this case the diagonal
elements of B are the corresponding eigenvalues.

We have the following equivalent statement.

Alternative form of theorem 4: An n-square matrix A is similar to a diagonal matrix B if
and only if A has n linearly independent eigenvectors. In this case the diagonal
elements of B are the corresponding eigenvalues.
In the above theorem, if we let p be the matrix whose columns are the 𝑛 independent
eigenvectors of A, then B =P-1 AP.
1 2
Example 4: Consider the matrix =( ) . By example 3, A has two independent
3 2
2 1 2 1
eighenvectors and . Set p = ( ) and so P -1 = 1/5 1/5
3 −1 3 −1
3/5 -2/5 .
Then A is similar to the diagonal matrix B=P-1 AP= 1 1
5 5 1 2 2 1
3 -2 3 2 3 -1
5 5
=4 0

0 -1

As expected, the diagonal elements 4 and -1 of the diagonal matrix B are the eigenvalues

corresponding to the given eigenvectors.

4.6. Characteristic Polynomial

Consider an 𝑛 –square matrix A over a field K: A = 𝛼 11 𝛼 12....… 𝛼 1𝑛
𝛼 21 𝛼 22…......𝛼 2𝑛
...…………..……..
𝛼𝑛1 𝛼𝑛2.....𝛼𝑛𝑛

36
 The matrix t I n -A, where In is the 𝑛 - square identify matrix and t is an inter
determinant, is called the characteristic matrix A: tIn -A= 𝑡 − 𝛼11 −𝛼 12 ..... 𝛼 1𝑛
−𝛼12 𝑡 − 𝛼 22 .....𝛼 2𝑛
..…………..……............
−𝛼𝑛1 −𝛼𝑛2...𝑡 − 𝛼𝑛𝑛

Its determinant A(t) = det (t In- A) which a polynomial in T, is called the

characteristic polynomial of A. We also call A(t) = det (tIn – A) = 0 the characteristic
equation of A.
Now each term in the determinant contains one and only one entry from each row and
from each column; is hence the above characteristic polynomial is of the form:
A(t)=(t –a11) (t–a22).......(t– 𝛼𝑛𝑛) + terms with at most 𝑛 -2 factors of the form t
– 𝛼𝑖𝑖 accordingly,
A(t) =tn – (𝛼 11 + 𝛼22+..................+ 𝛼𝑛𝑛 ) tn-1 + terms of lower degree.
Example 5: The characteristic polynomial of the matrix A = 1 3 0 is
-2 2 -1
4 0 -2
(t) = tI – A = t-1 -3 0 = t –t + 2t +28
3 2

2 t-2 -1
-4 0 t+2
As expected, (t) is a monic polynomial of degree 3. (A polynomial is monic if
its leading coefficient is I).
Now we state on of most important theorems in linear algebra.
Cayley Hamilton Theorem:
Every matrix is a zero of its characteristic polynomial.
1 2
Example 6: The characteristic polynomial of the matrix A= ( ) is (t)= tI-A = t2–1 -2
3 2
3 t- 2
= t2 – 3t – 4.
As expected from the cayley-Hamilton theorem, A is a zero of (t):
(A) = 1 2 2 3 1 2 -4 1 0 =0 0
3 2 3 2 0 1 0 0
Theorem 5: If the characteristic polynomial (t) of an 𝑛 -square matrix A is a product of
distinct linear factors: (t) = (t −𝛼 1) (t—𝛼 2).....(t – 𝛼𝑛), i.e. if a,.... 𝛼𝑛, are distinct roots
of (t) then A is similar to a diagonal matrix whose diagonal elements are the 𝛼i .
Furthermore, using the fundamental theorem of algebra (every polynomial over C has a
root and the above theorem we obtain.

Corollary 5: Let A be an 𝑛 –square matrix over the complex field C. Then A has at least
one eigenvalue.
37
 Example 7: Let A = 3 0 0 .Its characteristic polynomial is (t)= t-3 0 0
0 2 -5 0 t-2 5 = (t -3) (t+2).
0 1 -2 0 -1 t+2
Theorem 6: Similar matrices have the same characteristic polynomial.

4.7. 4.7 Minimum Polynomial

Let A, be an 𝑛 -square matrix over a fields K. Observe that there are nonzero
polynomials f(t) for which f(A) =0; for example, the characteristic polynomial of A.
Among these polynomials we consider those of lowest degree and from them
we select one whose leading coefficient is 1,.i.e which is monic. Such a polynomial
(m(t) exists and is unique. We call it the minimum polynomial of A.
Theorem 6: The minimum polynomial m(t) of A divides every polynomial which has
A as a zero. In particular m(t) divides the characteristic polynomial (t) of A.
Theorem 7: The characteristic and polynomial of a matrix A have the same irreducible
factors.
Theorem 8: A scalar 𝜆 is an eigenvalue for a matrix A if and only if 𝜆 is a root of
the minimum polynomial of A.
Example 7: Find the minimum polynomial m(t) of the matrix A = 2 1 0 0
0 2 0 0
0 0 2 0
0 0 0 5
The characteristic polynomial of A is (t) = t I –A = (t – 2)3 (t – 5). By theorem 7 both t-
2 and t -5 must be factors of m(t). But by theorem 6,m(t) must divide (t); hence m(t)
must be one of the following three polynomials :
m1(t) = (t -2) (t-5), m2(t)(t- 2)2 (t – 5), t m3 (t)=(t-2)3 (t -5).
Accordingly,m2 (t)=(t-2)2 (t – 5) is the minimum polynomial of A.
Example 8: Let A be a 3 by 3 matrix over the real field R. We show that a cannot be a
zero of the polynomial f(t) = t2 +1. By the Cayley Hamilton theorem, A is a zero of

its characteristic polynomial (t). Notice that (t) is of degree 3; hence it has at
least one real root.

38
 4.8. Characteristic and Minimum Polynomial of Linear Operators
Now suppose T:V V is a linear operator on a vector space V with finite dimension we
define the characteristic polynomial (t) of T to be the characteristic polynomial of any matrix
representation of T by theorem 6, (t) is independent of the particular basis in which the
matrix representation is computed. Note that the degree of (t) is equal to the dimension of V.

Theorem 9: The geometric multiplicity of an eigenvalue 𝜆 does not exercise its algebraic multiplicity.
Example 9: Let V be the vector space of functions which has (sin𝜃 cos 𝜃) as a basis and let D be the
Differential operator on V. Then D (sin 𝜃)=Cos 𝜃 = 0 (sin 𝜃 )+ 1 (cos 𝜃). 𝐷 (𝑐𝑜𝑠𝑡 𝜃) = −sin 𝜃=
-1(sin 𝜃)+0(cos 𝜃)
The matrix A of D in the above basis is therefore
A= [D] = 0 1 . Thus
-1 0
𝑡 −1
Det (tI-A)= | | = t2 + 1, and the characteristic polynomial of is (t) = t2+1.
1 𝑡
On the other hand, the minimum polynomial m(t) of the operator is defined independently of the theory
of matrices, as the polynomial of lowest degree and leading coefficient 1 which has T as a zero.
However for any polynomial f(t), f(T) =0 if and only if f(A)=0 where A is any matrix representation of
T. Accordingly, T and A have the same minimum polynomial. All the theorems in this unit on the minimum
polynomial of a matrix also hold for the minimum polynomial of the operator T.

EXERCISE
1 − 2
1. Find f(A) where ( ) and f(t) =t2 – 3t + 7.
4 5
1 4
2. Show that A =( ) is a zero of f(t)=t2 - 4t-5.
2 3
1 4
3. Let A =( ).
2 3
(i) Find all eigenvalues of A and the corresponding eigenvectors
(ii) Find an invertible matrix P such that P-1 AP is diagonal.

4. For each matrix, find all eigenvalues and a basis of each eigenspace:

39
 (i) A= 1 -3 3 , (ii) B= -3 1 -1
3 -5 3 -7 5 -1
6 -6 4

5. Let A = 3 -1 and B = 1 -1 . Find all eigen values and the corresponding eigenvectors
1 1 2 -1
Of A and B viewed as matrices over (i) the real field R, (ii) the complex field C.

6. Find all eigenvalues and a basis of each eigenspace of the operator T: R3 R3 defined by
T= x,y,z = 2x+y, y – z, 2y + 4z .

7. Let A = 1 2 3 . Is A similar to a diagonal matrix 2 if so, find such matrix.

0 2 3
0 0 3
8. For each matrix find a polynomial having the matrix as a root;
(i) A=2 5 , (ii) B= 2 -3 (iii) C= 1 4 3
1 -3 7 -4 0 3 1
0 2 -1.
9. Find the minimum polynomial
m(t) of A = 2 1 0 0
0 2 0 0
0 0 1 1
0 0 -2 4

10. Show that A= 1 1 is not diagonalizable table.

0 1

11. Let f(t) =2t2 – 5t +6 and g(t) = t3 _2t2+t +3. Find f(A), g(A), f(B) and g(B) where A = 2 -3 and
5 1
B= 1 2
0 3 .

12. Let B= 8 12 0 . Find a real matrix A such that B=A3.

0 8 12
0 0 8

13. For each matrix, find all eigenvalues and linearly independent eigenvectors :
(i) A= 2 2 (ii) B= 4 2 (ii) c= 5 -1

40
 1 3 , 3 3 1 3
14. For each matrix, find All eigen values and a basis for each eigenspace .
(i) A= 3 1 1 , (ii) B= 1 2 2 (iii) C= 1 1 0
2 4 2 1 2 -1 0 1 0
1 1 3 -1 1 4 0 0 1 .
15. For each matrix , find a polynomial for which the matrix is a root:

(i) A= 3 -7 , (ii) B= 5 -1 (iii) C= 2 3 -2

4 5 8 3 0 5 4 .
1 0 -1
16. Let A= a b be a matrix over the real field R. Find necessary and sufficient condition on
c d
a, b, c and d so that A is diagonalizable, that is has two linearly independent eigenvectors.

17. Show that a matrix is diagonalizable if and only if its minimum polynomial is a product of distinct linear
factors.

18. Let A= 1 1 0 and B= 2 0 0

0 2 0 0 2 2 . Show that A and B have different characteristic polynomials,
0 0 1 0 0 1
and so are not similar but have the same minimum polynomial.

41
5. REFERENCES
1. Pure mathematics 1 and 2, Backhouse and J.K. Houldsworth. P. T, 1985
2. Modern Mathematics for schools, Teachers Book 9, Second Edition, 1975
3. Mathematics for Scientific and Technical Students, Davies, H. G and Hicks, G.A, 1995
4. Theory and Problems of Linear Algerbra, Seymour, L. 1987.

42