SMA 212: LINEAR ALGEBRA I NOTES

1
 SMA 212: LINEAR ALGEBRA I

Course Outline
Pre-requisite

SMA 112: Basic Mathematics

Purpose of the Course

To introduce the learner to basic matrix algebra and vectors together with their applications.

Expected Learning Outcomes

By the end of the course, the learner should be able to:

1. Use various matrix methods to solve systems of linear equations.

2. Apply concepts of dot product and cross product of vectors in finding the equations of
lines and planes in  3 .
3. Establish whether a given set of vectors is linearly dependent or independent.
4. Establish whether a given subset of a vector space is a subspace
5. Construct a basis and determine the dimension of given vector spaces.

Course Content (Revised)

Matrices: Basic operations, reduction to echelon form, Determinants and inverses of 2x2 and
3x3 matrices. Systems of linear equations: Solution by reduction to echelon, Gauss Jordan,
Gaussian elimination, Crammer’s Rule and Inverse matrix method. Vectors: co-ordinate
systems, vectors and their components, basic operations on vectors, dot products and cross
products. Planes and lines in R 3 . Linear combination, dependence and independence. Vector
spaces: Subspaces, spanning sets, basis and dimension.

Mode of Delivery

Class lectures, tutorials, discussions, problem solving, exercises.

Instructional Materials

Whiteboard/chalkboard, overhead projector, smart board.

Course Assessment

Continuous assessment tests: 40%

Final examination: 60%

2
Core Reading Materials

1. Hoffman, K. and Kunze, R. (1971). Linear Algebra (2nd Edition) by Hoffman and Kunze,
Pearson Publishers.
2. Seymour L. and Marc L.(2012). Schaum’s Outline Series of Linear Algebra (5th Edition),
MacGraw-Hill Publishers.
3. Fraleigh, B. and Beauregard, R. (1994). Linear Algebra (3rd Edition), Pearson Publishers.
4. Sharma, J. and Vasishtha, R. (1982). Linear Algebra(2nd Edition), Krishna Prakashan
Media.

3
 MATRICES

Matrices are used as shorthand for keeping essential data arranged in rows and columns i.e
matrices are used to summarize data in tabular form.

Definition: A matrix is an ordered rectangular array of numbers, usually enclosed in

parenthesis or square brackets. Capital (Upper – case) letters are used to denote matrices.

Order of a Matrix
The size of a Matrix is specified by the number of rows (horizontal) and the number of
columns (vertical).

A general matrix of order m  n is

   aij 
 a11 a12 a1n 

a a a
A   21 22 2n 
i  ith row
 
  j  jth column
 am1 am 2 amn 
Square matrix
A Square Matrix is a one with the same number of rows and columns i.e
n  n matrix.
Two matrices are of the same size if they have the same order.

Vector
A vector is a matrix with one row (1  n) or one column (n  1). A row vector is of the form
1  n, and a column vector is of the form m  1.
(111111)-row vector
1
 
1
 1  -column vector
 
1
1
 

Zero matrix
A zero matrix of order m  n is the matrix with a ij  0 i  1,  m, j  1 n .
Similarly we talk of zero rows and column vector;

0
 
0
0  0,0, 0 or   .
 
 
0
Diagonal matrix
A diagonal matrix is a square matrix whose every entry is a zero except the entries in the
main (or principal or leading) diagonal.
EG:
4
 1 0 0 
 
A   0 3 0  is a diagonal matrix.
 0 0 2 
 
In particular, an identity matrix is a diagonal matrix whose every diagonal entry is a zero.
EG:
1 0 0
 
I   0 1 0  is a 3  3 identity matrix.
0 0 1
 
Triangular matrix
An upper triangular matrix is a square matrix whose every entry below the main diagonal is a
zero.
EG:
1 2 3 4
 
 0 3 1 0
A is a 4  4 triangular matrix.
0 0 4 5
 
0 0 0 7
Similarly, a lower triangular matrix is a square matrix whose every entry above the main
diagonal is a zero.

Basic operations on matrices

Equality of Matrices
Two matrices A and B are said to be equal if they have the same order (size) m  n and
ai j  bi j i and j.

Example
Given that
 x  y 2r  s   4 6 
   , solve for x, y, s and r.
 x  y s  r   2 3

Solution:
Since
 x  y 2r  s   4 6 
   , then
 x  y s  r   2 3

x y 4 i)
2s  r  6 ii)
x y 2 iii)
sr 3 iv)
From i) and iii), x  3, y  1
From ii) and iv), s  3, r  0 .

5
Addition and Subtraction of matrices
This is performed on matrices of the same order (size). Let A and B be m  n matrices.

    
A + B = aij  bij  aij  bij  cij    , a m  n  matrix
Scalar Multiplication
This is performed on any matrix and the resulting matrix is the same size. Let A be any
matrix, then
c A  c  aij    caij  , where c is a scalar.
Each entry is multiplied by same scalar.

 
Dot product: let a and b be any two vector of size n (matrices with a single column or
row).
     
a =  a1 , a2 , an  , b   b1 , b2 , bn  . The dot product of a and b denoted a b , is given by
  n
a b  a1b1  a2b2   anbn   a i bi .
i 1
The dot product is also scalar product.
Dot product of a row and column vector of order n is given by,
 b1 
  n  
b
 a1 , a2 an   2   a1b1  a2b2   anbn   ai bi  a  b
i 1
 
 n
b

EG.
(2, 2, 1) . (1, -7, 1)= 2 1  2  (7)  11  2  14  1  11

Matrix Multiplication:
 
Let A   aik  be an m  n matrix, and B  bkj an n  s matrix. The matrix product AB is
the m  s matrix C   cij  , where c ij the dot product of the ith row of A and the jth column
of B.

   
n
i.e. AB  C ,  aik  bkj = cij ; C ij  Ai  B j  a
k 1
ik bkj

Remark:
1. Let A  m  n  , B  s  r  be two matrices .
C  AB exists iff n  s and C is m  n
C  BA exists iff r  m and C is s  n
2. It’s possible for AB to be defined while BA is not defined i.e. matrix multiplication is not
commutative.

Examples
6
 1  2 3   3 0 2
1. Let A    and B    . Then
 4 5  6  7 1 8

 1  3  2  0 3  2   4  2 5
A  B      
 4  7 5  1  6  8   3 6 2

 3  1 3   2  33   3 6 9 
2. 3A     
3  4 35 3   6  12 15  18
 2 4 6   9 0 6   7 4 0 
3. 2 A  3B     
 8 10 12   21 3 24   29 7 36 

1 2 1 1
4. If C    and D    , then
3 4 0 2
 1 2  1 1   1 1  2  0 1 1  2  2   1 5 
CD      
 3 4  0 2   3 1  4  0 3 1  4  2   3 11
and

 1 1  1 2   1 1  1  3 1  2  1  4   4 6 
DC      
 0 2  3 4   0 1  2  3 0  2  2  4   6 8 
The above example shows that matrix multiplication is not commutative, i.e.the products AB
and BA of matrices need not be equal.

Assessment Questions

1 3   2 0  4
1. Let A    and B    . Find (a) AB , (b) BA
 2  1 3  2 6 

1  2 0 
2. Given A  2,1 and B    , find (a) AB , (b) BA
 4 5  3 

 2  1
  1  2  5
3. Given A   1 0  and B    , find (a) AB , (b) BA
3 4   3 4 0 
 

Transpose of a matrix
The transpose of the matrix A is matrix B  AT , such that bi j  a ji i.e. the rows become
columns and vice versa.

7
Example 2.1
Find AT and  AT  given that,
T

2 1 3 5
A .
 4 2 1 2 
Solution
2 4 
 
1 2  2 1 3 5
a) AT   A 
T
b) T
 .
3 1   4 2 1 2 
 
5 2 

Properties of Matrix operations

1. A  B  B  A - Addition is commutative
2.  A  B  C  A  B  C  - Addition is associative
3. A  O  O  A  A, O is the identity for addition
4.   A  B  A  B , left distributive law.
5.    A  A  A , right distributive law
6.   A   A , associativity of scalar multiplication
7.  AB  A   scalar pull through
8.  A B C  ABC  - associativity of matrix multiplication
9. 1n A  A , AI m  A , A  n  m  - identity for matrix multiplication
10. AB  C   AB  AC , left distributive law
11.  A  B C  AC  BC, right distributive law
12. A T   A
T

13.  A  B T  AT  B T - transpose of the sum=sum of transpose

14.  AB T  B T AT - transpose of product = product of transpose
 AB   B  A  .
NOTE
Proof of most of these properties involves routine computations with components and entries.

Example 2.2
Show that  A  B  A  B   A 2  B 2 iff AB  BA i.e A & B commute

Solution:
 Suppose  A  B  A  B   A 2  B 2 and show that A & B commute.
A 2  B 2   A  B  A  B 
 A A  B  B A  B (by left distributive law)
 A 2  AB  BA  B 2
Hence AB  BA  0 ; AB  BA and so A & B commute.
 Assume A & B commute & show that  A  B  A  B   A 2  B 2

8
Since A & B commute AB  BA , thus
 A  B A  B  A A  B  B A  B
 A 2  AB  BA  B 2
but AB  BA;  A 2  AB  AB  B 2  A 2  B 2

Assessment Questions
1 0 1 0
 
1. Find the transpose A of the matrix A   2 3 4 5 
t

 4 4 4 4
 

1 2 0
2. Let A    . Find (a) AA , (b) A A
t t

3  1 4

9
DETERMINANTS OF 2 2 AND 3 3 MATRICES
Determinants are defined on square matrices only.

Determinant of 2  2 Matrices

a a12 
Let A   11  . Then,
 a21a22 
a a12
det A  11  a11 a22  a12 a21 .
a21 a22
The determinant is a scalar.

Example 3.1
 
1. Find the value of  such that  0.
3 2
Solution
 
     2   3   2  5  0
3  2
     5  0    0 or   5 .
1 x 1
2. Let  1 find x.
2  2x 2

x 3
3. Let  4 . Find x.
2 2x  1

Determinant of 3  3 Matrices

Method 1
+ - +
 a11 a12 a13 
 
Let A   a21 a22 a23 
a a33 
 31 a32 , then

+ - +
a11 a12 a12
a22 a23 a21 a23 a21 a22
det A  a21 a22 a23  a11  a12  a13
a32 a33 a31 a33 a31 a32
a31 a32 a33
 a11  a22 a33  a23a32   a12  a21a33  a23 a31   a13  a21a32  a22a31 

10
  a11 a 22 a33  a11 a 23 a32  a12 a 21 a33  a12 a 23 a31  a13 a 21 a32  a13 a 22 a31
 a11 a 22 a33  a12 a 23 a 31  a13 a 21 a32  a11 a 23 a32  a12 a 21 a 33  a13 a 22 a31

Note: This is a sum of 6 products, 3 positive and 3 negative. Each product has exactly one
factor from each row and column.

Example 3.2

3 1 2
1. Evaluate  1  1 4
 2 1 1
Solution:
3 1 2
1 4 1 4 1 1
1 1 4  3 1 2
1 1 2 1 2 1
2 1 1
 3[ 11  1 4 ]  1[1(1)  8]  2[1  2]  9  7  2  0

1 3 1
2. Evaluate 2 1 2
1 1 2

Solution:
1 3 1
1 2 2 2 2 1
2 1 2 =1 3 1  14  32  1 3   5
1 2 1 2 1 1
1 1 2

Method 2

Determinant of a 3  3 matrix can also be obtained by the following elementary scheme;

+ + + - - -

a11 a12 a13 a11 a12

a21 a22 a23 a21 a22

a31 a32 a33 a31 a32

11
det A  a11a22 a33  a12 a23a31  a13a21a32  a13 a22 a31  a11a23 a32  a12 a21a33 .

Example 3.3

Find det A given that

 2 3 4 
 
A   1 4 2
 3 10 1 
 

Solution + + + - - -

2 -3 -4 2 -3

-1 4 2 -1 4

3 10 1 3 10

det A   2  4 1    3  2  3    4    1 10     4   4  3   2  2 10     3   1 1

 8  18  40  48  40  3
 35.

Assessment Questions
1. Evaluate the determinant of each matrix:

 3  2 a b a 
(a)   , (b)  
4 5   b a  b 

3 2 ab a
(i)  3  5   2   4  23 (ii)  a  b a  b   a  a  b 2
4 5 b ab

k k
2. Determine those values of k for which  0.
4 2k

k k
 2k  4k  0 , or 2k k  2  0 .Hence k  0 ;and k  2 .That is, if k  0 or k  2 ,
2

4 2k
the determinant is zero.
3. Compute the determinant of each matrix:

12
  1 2 3 2 0 1 
   
(a)  4  2 3  (b)  4 2  3 
 2 5 1 5 3 1 
   

Properties of determinants
1. det (AB) = det A . det B

2. If one row or column of A is multiplied by scalar r to get B, det (B) = r det A

1 2 
det A  det    3 2 1
1 3 

3 6
det B  det    9  6  3  3det A
1 3

3. If row i = 0 or column j = 0, det A =0.

4. For an n  n matrix, det (r A) = r n det A

1 2 
det A  det    3 2 1
1 3 

 5 10 
det C  det 5 A  det    75  50  25  5 det A
2

 5 15 

5. If two rows or columns of A are identical, det A = 0.

1 2 
Eg det    22  0.
1 2 

6. If one row or column is a scalar multiple of another, det A = 0.

7. Adding a scalar multiple of a row or a column to another row or column respectively

leaves the determinant unchanged.

1 2  R2  2 R1  1 2 
   
1 2   1  2.1  3 2  2.2  6 

8. det A  det A
T

13
 1 2 
det A  det    3 2 1
1 3 
 1 1
det AT     3  2  1  det A .
 2 3

Example 4.1

Let A and B be 3  3 matrices with det A  3 and det B  6 . Find

a) det 2A b) det AB

Solution

a) det 2 A  23 det A  8  3  24

b) det AB  det A . det B  3   6   18.

Assessment Questions

Evaluate
1 2 3 5 0 9
1. 2  1 3 2. 2  1 3
1 0 1 1 0 1

4 3 2 4 3 2
3. 3  2 5 4. 3  2 5
2 4 6 1 2 3

14
 INVERSES OF 2 x 2 AND 3 x 3 MATRICES

Definition

A square matrix A is said to be invertible if there exists a matrix B with the property that

AB  BA  I ,

where I is the identity matrix.

1 0
EG. I2   
0 1

Such a matrix B is unique and is called the inverse of A, denoted by A1 .

Definition (Minors and cofactors)

Let A be a matrix of order n  n and aij be a general element of A. The minor of element aij
which we denote by M ij is the determinant of the  n  1   n  1 submatrix which is
obtained after deleting the i th row and j th column of A.

The number Aij   1

i j
M ij is called the cofactor of aij .

Example 5.1

 2 3 4 
Let A   0 4 2  , find the minor and cofactor of each element of A.
 1 1 5 
 

Solution:

Minors

4 2 0 2 0 4
M 11   18 M 12   2 M 13  4
1 5 1 5 1 1
3 4 2 4 2 3
M 21   11 M 22   14 M 23   5
1 5 1 5 1 1
3 4 2 4 2 3
M 31   10 M 32  4 M 33   8
4 2 0 2 0 4

Cofactors

15
A11  (1) 2 M 11  18 A12  (1)3 M 12  2 A13  (1) 4 M 13  4
A21  11 A 22  14 A 23  5
A31  10 A 32  4 A 33  8

Definition 5.3 (Classical Adjoint)

Classical adjoint of a square matrix A denoted by adj  A is the transpose of matrix of

cofactors of A.

Example 5.2

From Example 5.1 above;

Matrix of cofactors is

 18 2 4 
 
 11 14 5  .
 10 4 8 
 

Therefore,

 18 2 4   18 11 10 

T

   
adj  A    11 14 5    2 14 4  .
 10 4 8   4 8 
   5

Inverses of a 2 X 2 and 3  3 Matrices

Theorem

Let A be any n  n matrix such that A  0 , then

1
A1  adj  A .
A

In particular if n  2 ,

a b 1 1  d b 
A  and A   .
c d  ad  bc  c a 

Proof

Cofactors of A;
16
 A11  d  d A12   c  c
A21   b  b A22  a  a

The matrix of cofactors is

 A11 A12   d c 
  .
 A21 A22   b a 

Therefore
T
A A12   d b 
adj  A   11   .
 A21 A22   c a 

Hence

1 1  d b 
A1  adj  A   .
A ad  bc  c a 

Example 5.2

 2 5 1
Let A    , find A .
 1 3

Solution:

det A  2  3  5 1  6  5  1.

Therefore,

1  3 5   3 5 
A1    .
1  1 2   1 2 

Inverse of a 3 X 3 matrix
Recall:
 a11 a12 a13 
 
Let A   a21 a22 a23  , then
a a33 
 11 a32
T
 A11 A12 A13 
Adj (A) =  A21 A22 
A23  , the transpose of matrix of cofactors and
A A33 
 31 A32
1
A 1  adj(A)
det A
17
Example 5.3:
4 0 1
 
Find the inverse of A   2 2 0  .
3 1 1
 

Solution:
2 0 2 0 2 2
det A = 4 0 1  42  0  1 2  6  8  4  4 ; Therefore, A 1 exists.
1 1 3 1 3 1

Cofactors;
2 0 2 0 2 2
A11  2 A12    2 A13   4
1 1 3 1 3 1

0 1 4 1 4 0
A21    1 A22  1 A23    4
1 1 3 1 3 1

0 1 4 1 4 0
A31   2 A32   2 A33  8
2 0 2 0 2 2
Therefore
 2 2 4   2 1 2 
T

adj ( A)   1 1 4    2 1 2  .
 
 2 2 8   4 4 8 
Hence
 2 1 2   1 2 1
4  12 
adj ( A)   2 1 2     12 14 1 
1 1
A1  2 
det A 4
 4 4 8   1 1 2 

5.2.1.1 Properties of Inverses

1. AA1  A1 A  I
 AB 
1
2.  B 1 A1

A  A 
T 1
1 T
3.

A   A
1
1
4.
1
5. A .
A1

18
Assessment Questions
Find A 1 if
 2 3  4
 
1. A =  0  4 2 
 1 1 5 
 

19
 REDUCTION OF A MATRIX TO ECHELON FORM

Echelon form of a matrix

Definition
A matrix A   aij  is an echelon matrix, or is said to be in echelon form, if the number of
zeros preceeding the first non-zero entry of a row increases row by row until only zero rows
remain (if possible).

The leftmost non-zero entry in a non-zero row of an echelon matrix is called the leading
entry of that row.

Examples
1. The following matrices are in echelon form;
1 2 3 4
 
2 3 2 0 4 5 6  0 1 4 2
  1 2 3
0 0 7 1 3 2 1    0 0 7 3 
A , B  0 0 4, C  .
0 0 0 0 6 2 2 0 0 0 0 0 0 9
    0
0 0 0 0 0 0 1 0 0 0
 
0 0 0 0 
Note: The leading entries of A are 2, 7, 6 and 1.

2. The following matrix is not in echelon form;

 2 1 4 3 
 
0 1 3 2 
E  .
 0 2 3 1 
 
 0 2 6 4 

 We perform the following operations on rows of a matrix when reducing it to echelon

form ;
1. Interchange the i th row and j th row
Ri  R j .
2. Multiply i th row by a non-zero constant
Ri  kRi , k  0 .
3. Add a scalar multiple of row j to row i
Ri  Ri  kR j

In actual practice, we apply operations 2 and 3 in one step;

Ri  kRi  k ' R j , k  0 .
These operations are called elementary row operations.

20
Example 6.1
1 2 3 1
 
2 1 2 0
1. Reduce  to echelon form.
 1 2 1 1
 
 1 1 3 3
Solution
1 2 3 1  R2  R2  2 R1  1 2 3 1  1 2 3 1 
     
2 1 2 0   0 3 4 2    0 3 4 2 
 1 2 1 1  R3  R3  R1  0 0 2 2  R4  R4  R2  0 0 2 2 
     
 1 1 3 3  R4  R4  R1  0 3 0 2  0 0 4 4 

1 2 3 1 
  
 0 3 4 2 
R4  R4  2 R3  0 0 2 2 
 
0 0 0 0 
 1 2 3 1
2. Reduce  2 1 2 2  to echelon form.
3 1 2 3 
 

Solution:

 1 2 3 1 R2  R2  2 R1  1 2 3 1  1 2 3 1 
      
 2 1 2 2    0 3 4 4  R  3R  7 R  0 3 4 4 
 3 1 2 3  R  R  3R  0 7 7 6  3 3 2
7 10 
  3 3 1   0 0

 3 4 1 1 
3. Reduce  1 1 3 1  to echelon form.
 4 3 11 2 
 
Solution:
 3 4 1 1   1 1 3 1  R2  R2  3R1  1 1 3 1 
  R2  R1    
 1 1 3 1    3 4 1 1    0 7 10 2 
 4 3 11 2   4 3 11 2  R  R  4 R  0 1 1 2 
    3 3 1

21
  1 1 3 1   1 1 3 1 
R2  R3   R3  R3  7 R2  
0 1 1 2   0 1 1 2 
     0 0 3 12 
 0 7 10 2   

Reduced row-echelon form

Definition: An echelon matrix is said to be in reduced row form (canonical form) if:
1. The leading entry in each non-zero row is 1.
2. Each leading 1 is the only non-zero entry in its column.

EG: The following matrices are in reduced row (canonical form);

1 0 0 7 
0 1 0 3  1 0 0 0 7 6 
  0 1 0 0 2 1 

A 0 0 1 5  B 
  0 0 1 0 3 5 
0 0 0 0   
0 0 0 0  0 0 0 1 2 9 
NOTE:
Every matrix is row-equivalent to one and only one reduced row form.

Examples
1 2 1 4 
 
1. 1. Reduce  3 8 7 20  to reduced row-echelon (canonical) form.
 2 7 9 23 
 
Solution:
 1 2 1 4  R2  3R1  1 2 1 4  1 1 2 1 4  1 2 1 4
    2   R3  3R2  
 3 8 7 20    0 2 4 8  R2  0 1 2 4    0 1 2 4 
 2 7 9 23  R  2 R  0 3 7 15    0 3 7 15  0 0 1 3
  3 1     

R1  R3  1 2 0 1 1 0 0 5 
  R1  2 R2  
 0 1 0 2   0 1 0 2 
 
R2  2 R3  0 0 1 3  0 0 1 3 


3 4 1 1
 
2. Reduce  1 1 3 1  to (canonical) form.
 4 3 2 
 11

Solution:

22
 3 4 1 1 1 1 3 1  R2  3R1  1 1 3 1
     
 1 1 3 1  R1  R2  3 4 1 1  0 7 10 2  R2  R3
 4 3 11 2  4 3 11 2  R3  4 R1  0 1 1 2 
 
 1 1 3 1  R1  R2  1 0 2 1  1 1 0 2 1  R1  2 R3
     3 R3  
 0 1 1 2    0 1 1 2  0 1 1 2  
 0 7 10 2  R3  7 R2  0 0 3 12   0 0 1 4  R2  R3
 
1 0 0 7 
 
 0 1 0 6 
 0 0 1 4 
 

Inverse of a matrix (row reduction method)

The inverse of a matrix a can be found using row reduction to echelon form of the augmented

matrix  A I  to get I A1 . 
Note:
Augmented matrix is a matrix of the form
B   C D  , where C and D are matrices with the same number of rows.

Example:
4 0 1
 
Find the inverse of A   2 2 0 
3 1 1
 

Solution:
 4 0 1 1 0 0  2 R2  R1 4 0 1 1 0 0
    R3  R2
 2 2 0 0 1 0   0 4 1 1 2 0  
 3 1 1 0 0 1  4 R  3R  0 4 1 3 0 4 
  3 1  
1 R
8 1
4 0 1 1 0 0  2 R1  R3  8 0 0 4 2 4 
    1 R
 0 4 1 1 2 0   
0 8 0 4 2 4 8 2
 
0 0 2 2 2 4  2 R2  R3  0 0 2 2 2 4  

1 R
2 3
1 0 0 12 1
4  12 
 
 0 1 0  12 14 12  .
0 0 1 1 1 2 

Therefore,

23
  12 1
4  12 
 1 
A1    12 14 2 
 1 1 2 
 

Assessment Questions
1  2 1 
 
1. Reduce the matrix  2 1  1 to echelon form
7  4 1 
 

2. Determine if the following matrices are in echelon canonical form or not.

1 0 0 0 0
1 0 0 1  7 5 2 0   
     0 0 1 0 6
0 0 1 3 0 3 1 1
a)  b)  c)  0 1 0 0 3 
0 0 0 0  0 0 4 6   
     0 0 0 1 0
0 0 0 0 0 0 0 2  0 0 0 0 0
 
 5 8 7 3 
 
0 0 2 1 
d)  0 0 0 0 
 
0 0 0 4 
0 0 0 0 
 

 2 2 1 3 1 
 
3. Reduce the matrix  2 1 1 7 3  to echelon and canonical forms (row
 7 4 1 4 2 
 
reduced echelon form).

24
SOLUTION OF SYSTEMS OF LINEAR EQUATIONS (GAUSS-JORDAN METHOD)

Introduction to Solution of systems of linear equations

Definition
A linear equation is an equation of the form
a1 x1  a 2 x 2    a n x n  b
where a1 , a2 , , an , b are constants while x1 , x2 , , xn are the variables(unknowns) to be
determined. A solution to this equation is an n-tuple c1 , c 2 ,  , c n  of numbers such that
a1 c1  a2 c2   an cn  b.
Example:
For 5 x1  3x2  6 x3  4 x4  10 ; x1  3 ; x2  5 ; x3  3 ; x4  2 . i.e. (3, 5, 3, 2) is a solution.

Definition 7.2
A system of linear equations is a collection of several linear equations. A solution to a system
of linear equations in n variables is an n-tuple c1 , c 2 ,  , c n  that satisfies each of the linear
equations in the system.
A linear system is said to be consistent if it has at least one solution and inconsistent if it has
none.

Methods of Solution

Elimination Method
The strategy is to eliminate one variable at a time by performing one of the following
elementary operations;

1. Interchange two equations

2. Multiply one equation by a non-zero constant.
3. Add a constant multiple of one equation to another equation.

Example 7.1:
Solve by elimination method
x1  x 2  x3  5  (1)
x1  2 x 2  3x3  10  (2)
2 x1  x2  x3  6 (3)

Solution:
Eliminate x1 from equations (2) and (3),
x1  x2  x3  5 (1)
(2) – (1) x 2  2 x3  5 (4)
(3)– 2 (1)  x2  x3  4 (5)

Eliminate x2 from equation (5),

x1  x2  x3  5

25
 x2  2 x3  5
(5) + (4) x3  1

Substitute x3  1 in equation (4)

x2  2  5  x2  3
Substitute x3  1 and x2  3 in equation (1),
x1  3  1  5  x1  1
x1  1, x2  3, x3  1
Solution:

The elimination method is very tedious especially when there are many variables. We use a
much more organized elimination method in matrix form called Gauss-Jordan elimination
method.

Gaussian Elimination Method

Given a linear system
a11 x1  a12 x 2    a1n x n  b1
a21 x1  a22 x2   a2 n xn  b2

a n1 x1  a n 2 x 2    a nn x n  bn

The system can be represented in matrix form

 a11 a12 a1n   x1   b1 
     
 a21 a22 a2 n   x2   b2 

     
     
 an1 an 2 ann   xn   bn 

Or simply Ax = b, where A is the n  n is the coefficient matrix and

x   x1 , x2 , , xn  , b   b1 , b2 , , bn  .
T T

The Gaussian elimination method involves reducing the augmented matrix  A b  to echelon
form and then doing back substitution.

Example 7.2:

Solve the system using Gausian elimination Method;

x1  2 x2  x3  4
3 x1  8 x2  7 x3  20
2 x1  7 x2  9 x3  23

Solution:
The Matrix form is,

26
 1 2 1  x1   4 
    
 3 8 7  x2    20  .
 2 7 9  x   23 
  3   
We reduce the augmented matrix to echelon form,

1 2 1 4  R2  3R1  1 2 1 4 1 2 1 4 1 2 1 4
    1 2 R2   R3  3R2  
3 8 7 20    0 2 4 8 0 1 2 4 0 1 2 4
2 7 9   
 23  R3  2 R1  0 3 7 
15  0 3 7


15  0 0 1 3 

Writing in equation form;

 1 2 1   x1   4 
    
 0 1 2   x2    4 
0 0 1  x   3
  3   

x1  2 x2  x3  4
x2  2 x3  4
x3  3
 x3  3
Doing back substitution,
x2  4  6  2
x1  4  4  3  5
Solution: x1  5, x2  2, x3  3 .

Remark
1. Suppose a linear system is in echelon form i.e. the augmented matrix  A b  is an
echelon matrix. Then those variables that correspond to columns containing leading
entries are called leading (dependent) variables, all other variables are called free
(independent) variables. E.g. in the previous example 7.2 above x1 , x2 and x3 are
all leading/dependent variables.
2. A system of linear equations may have a unique solution, many solutions or no
solutions;
Unique solution: The number of unknowns and equations are equal in the echelon
form of the system. E.g. Example 7.2 had a unique solution
Many solutions: Equations are less than the unknowns in the echelon form of the
system.
No solution: Last row of the echelon form of the system if of the form  00 0 a  0

Example 7.3
Solve the following system using Gaussian elimination method;
27
 x1  2 x2  3 x3  2 x4  x5  10
2 x1  4 x2  8 x3  3 x4  10 x5  7
3 x1  6 x2  10 x3  6 x4  5 x5  27

Solution:
 x1 
 
 1 2 3 2 1   x2   10 
    
 2 4 8 3 10   x3    7 
 3 6 10 6 5  x  
   4   27 
x 
 5
Reducing the augmented matrix to echelon form,
 1 2 3 2 1 10   1 2 3 2 1 10   1 2 3 2 1 10 
  R2  2 R1    
 2 4 8 3 10 7
R3  3R1 
0 0 2 1 8 13  R2  R3  0 0 1 0 2 3 
 3 6 10 6 5 
27  3   0 0 2 1 8 13 
 0 0 1 0 2 

 1 2 3 2 1 10   1 2 3 2 1 10 
   R3  
R3  2 R2  0 0 1 0 2 3  0 0 1 0 2 3 
 0 0 0 1 4   
 7   0 0 0 1 4 7 

The corresponding system is

x1  2 x2  3 x3  2 x4  x5  10
x3 + 2x5  3  Infinitely many solutions (equations less than unknowns)
x4  4 x5  7

The leading variables (dependent variables) are x1 , x3 , x4 while the independent (free)

variables are x2 and x5 . Let x2  s and x5  t , where .

Then the general solutions are;

x4  4 x5  7  4t  7

x3 =-2 x5  3  2t  3

28
x1  10  2 x2  3x3  2 x4  x5
 10  2s  3  2t  3  2  4t  7   t  5  2s  3t

i.e.
x1  5  2s  3t

x2  s

x3  2t  3

x4  4t  7

x5  t

Or

 x1   5   2   3 
       
 x2   0   1   0 
 x3    3   s  0   t  2  , s, t  .
       
 x4   7   0   4 
 x   0  0  1 
 5      

Eg Setting x2  s  1, x5  t  1 gives the solution: x1  4, x2  1, x3  5, x4  11, x5  1 .

Infinitely many other solutions can be obtained in the same way.

Gauss-Jordan Elimination Method

This method involves reducing the augumented matrix  A b  to row reduced echelon form
(canonical form). The solutions will be obtained directly; there is no need of doing back
substitution.

Example
Solve using Gauss-Jordan method
2 x1  4 x 2  6 x3  20
3x1  6 x 2  x 3  22
 2 x1  5 x 2  2 x3   18
Solution:
Matrix form
 2 4 6   x1   20 
     
 3 6 1   x2  =  22 
 2 5 2   x   18 
   3  
We reduce the augmented matrix to canonical form;

29
 2 4 6 20   1 2 3 10  R2  3R1  1 2 3 10 
  1 2 R1    
 3 6 1 22   3 6 1 22   0 0 8 8 
 2 5 2 18    2 5 2 18  R  2 R  0 1 4 2 
    3 1 

 1 2 3 10   1 0 11 14  1 0 1114  R1  11R3
  R1  2 R2    18 R3  
R 2  R3  0 1 4 2  0 1 4 2  0 1 4 2 
 0 0 8 8    0 0 8 8   0
     0 1 1  R2  4 R3
1 0 0 3 
 
 0 1 0 2 
0 0 1 1 
 
Therefore
 1 0 0  x1   3 
    
 0 1 0  x2    2 
 0 0 1  x   1 
  3   

Solution: x1  3, x2  2, x3  1

Remark 7.2
A system of linear equations is homogeneous if it can be written in the form Ax  0 , where
A is the coefficient matrix and 0 is the zero column vector. Such a system always has at least
one solution, namely x  0 (zero vector); this is usually called the trivial solution.
A non-trivial solution exists if and only if the system has at least one free (independent)
variable or if the determinant of the coefficient matrix is zero.

Assessment Questions
Solve the following system by Gauss Jordan (elimination) method.
1. x1  x 2  x3  2 x 4  1
2 x1  x2  3x4  0
 x1  x2  x3  x4  1
x2  x4  1

2. x1  x 2  x3  x 4  4
x1  2 x 2  x3  x 4  7
2 x1  x 2  x3  x 4  8
x1  x 2  2 x3  2 x 4  7

3. x1  x2  2 x3  x4  3

30
 x1  2 x2  x3  x4  2
x1  x2  x3  2 x4  1
2 x1  x2  x3  x4  4

4. x1  x2  2 x3  2
2 x1  3x2  x3  14
3x1  2 x2  x3  16
x1  4 x2  3x3  12

31
 CRAMER’S RULE AND INVERSE MATRIX METHOD

Cramer’s Rule
This method uses determinants to solve a linear system Ax  b provided determinant of A
(coefficient matrix) is nonzero.

Theorem 8.1(Crammer’s Rule)

Consider a system of n linear equations with n unknowns;

a11 x1  a12 x 2    a1n x n  b1

a21 x1  a22 x2   a2 n xn  b2

a n1 x1  a n 2 x 2    a nn x n  bn
Let  denote the determinant of the matrix of coefficient A   aij  . Also, let  i denote the
determinant of the matrix obtained after replacing the i th column of A by the column of
constant terms b   b1 , b2 , , bn  . Then the above system has a unique solution if and only if
T

  0 . In this case the unique solution is given by

  
x1  1 , x2  2 , , xn  n .
  

Example 8.1
Solve the following linear system using Crammer’s rule
 2 x1  3 x2  x3  1
x1  2 x2  x3  4
 2 x1  x2  x3   3

Solution:

 2 3 1 1
   
A   1 2 1 and b   4  .
 2 1 1   3 
   
Then

2 3 1
2 1 1 1 1 2
 A  1 2 1  2 3 1  2 1  3  1  1 3  2  3  3  2.
1 1 2 1 2 1
2 1 1
Next,

32
 1 3 1
2 1 4 1 4 2
1  4 2 1  1 3 1  1  3  2  4
1 1 3 1 3 1
3 1 1

4
 x1 
2
2
2 1 1
4 1 1 1 1 4
2  1 4 1  2 1   2  1  5   6
3 1 2 1 2 3
2 3 1
6
 x2  3
2
2 3 1
2 4 1 4 1 2
3  1 2 4  2 3 1  4  15  3  8
1 3 2 3 2 1
2 1 3
8
 x3  4
2

Inverse matrix method

Consider the system of linear equations represented in matrix form A x  b , where A is an
n  n matrix.
If det A  0 , then A 1 exists and
A 1  A x   A 1b  Ix  x  A 1b ,
which gives the solution directly. This method works only if det A  0 and gives the unique
solution
 x1   c1 
x   
 2   A1b   c2 
   
   
 xn   cn 

Examples:
1. Solve the linear system using the inverse matrix
x  2y  6
4x  3y  3
Solution:
1 2  x   6 1 2 1  3 2 
    ; A  , det A  3  8  5 ; A1    
 4 3  y   3  4 3 5  4 1 
 12 
 x 1 1  3 2   6  1  12    5 
   A b           .
 y 5  4 1   3  5  21  21 
 5 
Solution: x   12 5 , y  215

33
2. Solve the linear system using the inverse matrix
x1  3x2  2 x3  3
2 x1  4 x2  2 x3  8
x1  2 x2  x3  10
Solution:
 1 3 2   x1  3 1 3 2 
       
 2 4 2   x2  =  8  ; A   2 4 2 
 1 2 1  x  10   1 2 1 
   3    

We first find A1 . We use the reduction to echelon form method,

1 3 2 1 0 0  R2  2 R1  1 3 2 1 0 0 1 3 2 1 0 0
    R2  R3  
2 4 2 0 1 0    0 2 2 2 1 0   0 1 3 1 0 1 
 1 2 1 0 0 
 1  R3  R1  0 1 3 1 0 1   0 2 2
 2 1 0 
1 3 2 1 0 0  R1  3R2  1 0 7 2 0 3  4 R1  7 R3  4 0 0 8 7 2 
     
0 1 3 1 0 1   0 1 3 1 0 1   0 4 0 4 3 2 
 R2 
 0 2 2 2 1 0  R3  2 R2  0 0 4 0 1 2  4 R2  3R3  0
 0 4 0 1 2 

1 R  2 7 1 
4 1 1 0 0 4 2
1 R  0 1 0 1 3 1 .
4 2 4 2
1 R  0 0 1 0 1 1 
4 3 4 2

Therefore
 2 7 1 
 4 2
A  1 3
1 1 
 4 2 
 0 1  1 
 4 2
The solution is given by,
 2 7 1 
 x1   4 2  3   3 
  1  1   8  2 
 x2   A b   1  4
3
2     
x    10   3 
 3  0 1  1     
 4 2

Thus
x1  3, x2  2 , x3  3

34
Assessment Questions
Solve using Crammer’s rule.
1. 2x  5 y  2z  7
x  2 y  4z  3
3x  4 y  6 z  5

x1  3 x2  2 x3  3
2. 2 x1  4 x2  2 x3  8
x1  2 x2  x3  10

3. Solve the following three equations using inverse matrix method

x  y  2z  1 x  2y  z  4
x  2 y  z  2 4. 3 x  4 y  2 z  2
x  3y  z  5 5 x  3 y  5 z  1

35
 VECTORS

Co-ordinate Systems
Any point on the Cartesian plane (xy- plane) can be represented as a pair  x0 , y0  where x 0
is the x coordinate and y 0 is y-coordinate. The Cartesian plane is denoted by  2 .
Next, 3 consists of three planes, xy - plane, xz - plane and yz - planes. The x-axis, y-axis
and z – axes all meet at 90  (are perpendicular) at the origin (0, 0, 0). The x-axis consists of
points  x0 , 0, 0  , y-axis  0, y0 , 0  and z-axis  0, 0, z0  .
 n consists of all n-tuples consisting of real entries i.e.
n   a1 , a2 , , an  : ai , i  1, 2, , n .

Definition (Vector): They are quantities defined by magnitude and direction.

 a1 
 
 a2 
A column vector is an n-tuple of numbers written vertically   (or n  1 matrix)
 
 an 
 
 
If the numbers a1 , a2 , , an are real we have a real column vector. If they are complex we
have a complex column vector. The number a i in the ith slot is called the ith component.
A row n-vector is an n-tuple  a1 , a2 , , an  of numbers written horizontally.

The zero vector in  n is a vector of the form,

0   0, 0, , 0  .
~
The unit vector is a vector of length 1.
In  2 , the unit vector along the x-axis is denoted by i and along the y-axis is denoted by j ;
~ ~

i  1, 0  , j   0, 1 .
~ ~

Similarly, in 3 , the unit vector along the x-axis is denoted by i and along the y-axis is
~

denoted by j and along the z-axis by k~ ;

~

i  1, 0,0  , j   0,1,0  , k   0, 0,1 .

~ ~ ~

If  x, y, z   , then 3

 x, y, z   x i~  y j  z k~ .
~

Length/Magnitude of a Vector:
Let a   a1 , a2  be a vector in  2 . Then by Pythagoras Theorem,

a  a12  a22
~

36
In general, if a   a1 , a2 , , an  n , then
~

a  a12  a22   an 2

9.2.1.1 Vector Algebra

a) Equality of vectors
Let a   a1 , a2 , , an  , b   b1 , b2 , , bn  n , then
~ ~

 a1 , a2 , , an    b1 , b2 , , bn  if ai  bi , i  1, 2, ,n.

b) Addition & Subtraction

Let a   a1 , a2 , , an  , b   b1 , b2 , , bn  n , then
~ ~

a  b   a1 , a2 , , an    b1 , b2 , , bn    a1  b1 , a2  b2 , , an  bn  .
~ ~

c) Scalar multiplication
Let a   a1 , a2 , , an  n and k be any scalar, then
~

k a  k  a1 , a2 , , an    ka1 , ka2 , , kan 

~

Example 9.1
1. If u   4, 3 and v   2,3 , then
~ ~

u  v   4,  3   2, 3   2, 0 
~ ~

u  v   4,  3   2, 3   6, 6 
~ ~

2 v  2  2, 3   4, 6 
~

2. Solve for  and  if

  3, 4    1,  1  1, 2  .
37
Solution:
  3, 4    1,  1  1, 2 
  3 , 4     ,     1, 2 
 3   , 4     1, 2 
 3    1 i)
4    2 ii)

On solving,
  37 ,    2 7 .
Therefore
3  3, 4   2 1,  1  1, 2  .
7 7

9.2.1.2 Properties of Vector operations

Let a, b, c be n-row vectors,  ,  scalars, then
~ ~ ~

1. a b  b a Commutative law
~ ~ ~ ~

2.  a  b   c  a   b c 
~ ~ ~ ~ ~ ~
Associative law
3. 0 a  a 0  a Additive identity
~ ~ ~ ~ ~

~
   
4. a  a   a  a  0
~ ~ ~
Additive inverse

 
5.  a  b   a   b
~ ~ ~ ~
Left distributive law

6.      a~   a~   a~ Right distributive law

7.     a~     a~  Scalar multiplication associativity

   
8.   a =  a
~ ~
Scalar pull through
9. 1.a  a Multiplication identity
~ ~

Dot Products
Definition 9.1 (Algebraic definition of dot products)
Let a   a1 , a2 , , an  , b  b1 , b2 , , bn  n , then
~ ~

a  b  a1b1  a2b2   anbn .

~ ~

Example 9.2
a) 1, 2    2,3  1 2  2  3  8
b) 1, 2,3   1, 2,1  1  1  2  2  3 1  6
c)  2 i  j    3 i  j   2  3  1  1  5
 ~ ~  ~ ~
38
Definition 9.2 (Geometric definition of dot product)
If  is the angle between the vectors a and b , then
~ ~

a  b  a b cos  .
~ ~ ~ ~

Proof
(Exercise)
Hint: Use the Cosine Rule

Proof:
Let a   a1 , a2  , b   b1 , b2  2 and consider the figure below;
~ ~

By Cosine Rule,
2 2 2
a  b  a  b  2 a b cos 
~ ~ ~ ~ ~ ~

So
2 2 2
2 a b cos   a  b  a  b
~ ~ ~ ~ ~ ~

RHS
2 2 2
a  b  a  b  a1 , a2  b1 , b2  a1  b1 , a2  b2
2 2 2

~ ~ ~ ~


 a12  a22  b12  b22  a12  2a1b1  b12  a22  2a2b2  b22 
 a12  a22  b12  b22  a12  2a1b1  b12  a22  2a2b2  b22

39
 2a1b1  2a2b2

Therefore,

2 a b cos   2  a1b1  a2b2 

~ ~

 a b cos   a1b1  a2b2  a . b

~ ~ ~ ~

In general,
 a1 , a2 , , an . b1 , b2 , , bn   a  b  a1b1  a2b2 
~ ~
 anbn

Properties of the dot product:

1. a  b  b  a Commutative property
~ ~ ~ ~

~
 
2. a  b c  a b a c
~ ~
Distributive Property
~ ~ ~ ~

3.  r a   b  a   r b   r  a  b  Homogeneous property
~ ~ ~ ~ ~ ~

4. a 0  0
~ ~

5. i  j  j  k  i  k  0 , they are perpendicular therefore the angle between them is 90 0 .

~ ~ ~ ~ ~ ~

6. i  i  j  j  k  k  1 , the angle between them is 00

~ ~ ~ ~ ~ ~

7. a  b  0 if and only if a and b are perpendicular or at right angles or are orthogonal.

~ ~ ~ ~

8. The angle between a and b is given by

~ ~

a b
cos   ~ ~
.
a b
~ ~

2
9. a  aa  aa  a (Show using a vector in  n ).
~ ~ ~ ~ ~ ~

Example 9.3:
1. Find the angle between u  i  4 j  8k and v  5 i  4 j  3k .
~ ~ ~ ~ ~ ~ ~ ~

Solution:
 i  4 j  8k    5 i  4 j  3k 
u v ~   

~   ~ ~ 
cos   ~ ~ ~ ~

u v
~ ~
i  4 j  8k 5 i  4 j  3k
~ ~ ~ ~ ~ ~

40
 5  16  24 45 1
  
12  42  82 52  42  32 81 50 2
 1 
   cos 1    45 .
0

 2

2. Determine c such that a  3 i  4 j  2 k and b  c i  5 j  3 k are orthogonal.

~ ~ ~ ~ ~ ~ ~ ~

Solution:

We determine c such that a . b  0 . Now,

~ ~

a . b  (3, 4, 2). (c,5, 3)  0

~ ~

 3c  20  6  0
3c  26
 c  26
3

Exercise

Prove that the diagonals of a rhombus are perpendicular.

Cross Product/Vector products

These products are only defined in three dimensions.

Definition 9.3 (Algebraic definition of cross products)

The cross product of a   a1 , a2 , a3  and b   b1 , b2 , b3  , denoted by a  b is a vector
~ ~ ~ ~
orthogonal to both a and b defined by
~ ~

i j k
~ ~ ~
a2 a3 a1 a3 a1 a2
a  b  a1 a2 a3  i j k .
~ ~ ~ b2 b3 ~ b1 b3 ~ b1 b2
b1 b2 b3

Example 9: Let a   3 i  j  k  ; b   i  2 j  k 
 ~ ~ ~ ~ ~ ~ 

a) Find a  b and b  a
b) Show that a  b is orthogonal to a
c) Show that a  b is orthogonal to b

41
Solution:
a)
i j k
~ ~ ~

i) a  b   3 i  j  k    i  2 j  k   3 1 1  i (1  2)  j (3  1)  k (6   1)   i  4 j  7 k .
 ~ ~ ~  ~ ~ ~  ~ ~ ~ ~ ~ ~
1 2 1

i j k
~ ~ ~

ii) b  a   i  2 j  k    3 i  j  k   1 2 1  i (2  1)  j (1  3)  k ( 1  6)  i  4 j  7 k .
~ ~ ~   ~ ~ ~  ~ ~ ~ ~ ~ ~
3 1 1

 ab   ba  
Conclusion: Vector products are not commutative

b) a  (a  b)   3 i  j  k     i  4 j  7 k   3  4  7  0 and therefore a  b is orthogonal

 ~ ~ ~  ~ ~ ~

to a

   
c) b  (a  b)   i  2 j  k     i  4 j  7 k    1  8  7  0 and therefore a  b is orthogonal
~ ~ ~   ~ ~ ~

to b .

Definition 9.4 (Geometric definition of cross products)

a  b is a vector whose magnitude is a b sin  , where  is the angle between the vectors
~ ~ ~ ~

a and b . Now
~ ~

 
a  b  a b sin  u , where u is a unit vector perpendicular to both a and b .
~ ~ ~ ~ ~ ~

42
 -unit vector orthogonal to and

Properties of the Cross Product:

Let a , b , c be vectors,  ,  be scalars. Then
~ ~ ~

~
 
1. a  a  b  0 i.e a  b is orthogonal to a
~ ~ ~ ~ ~

2. a  b    b  a 
~ ~ ~ ~

3. a   b c    a  b    a  c 
~ ~ ~ ~ ~ ~ ~

 a b  c  a  c  b  c
~ ~ ~ ~ ~ ~ ~

5. a  a  0.
~ ~ ~

Remark
Two vectors a and b are parallel if a  kb for some scalar k.

Assessment Questions
1
 
1. Compute the length of the vector v   2 
 3
~

 
 2  4
. Show that the vectors u    and v    are orthogonal.
~
 4  ~
 2

43
 PLANES AND LINES IN 3
Lines in 3-dimensions spaces
A line in space is determined by a single point P0 on it and a vector a that determines the
~

direction of the line.

Let P0  x0 , y0 , z0  be a point on line L which is parallel to the vector a   a1 , a2 , a3  and

~

suppose P  x, y, z  is any point on L.

Line L

    
OP  OP0  P0 P and P0 P is parallel to a   a1 , a2 , a3  .

Then there exists a scalar such that

  
OP  OP0   a .
That is,

 x, y, z    x0 , y0 , z0     a1, a2 , a3  . 1)

Equation 1) is called the vector equation of the line L.

Equating the components of the vectors in 1), we obtain

x  x0   a1
y  y0   a2 2)
z  z0   a3

These are parametric equations of line L through point P0  x0 , y0 , z0  and parallel to vector
a   a1 , a2 , a3  .
~

44
Eliminating  from 2) gives,

x  x0 y  y0 z  z0
  3)
a1 a2 a3

Equation 3) is called the symmetric equation of line L.

Note: Vector  a1 , a2 , a3  is parallel to line L.

Example
Find the parametric and symmetric equations of the line passing through A (2, -3, 4) and
parallel to the vector a = (3, 5, -6).
Solution:
If R (x, y, z) is any point on this line, then

R (x, y, z)

A (2, -3, 4)

Line

OR  OA  AR,
OR  OA   a, where  is a scalar
~

Thus

 x, y, z    2,  3, 4    3, 5,  6 

is the vector equation of the line.

Now,
x  2  3
y  3  5
z  4  6

are the parametric equations of the line.

Eliminating  ,

45
 x2 y3 z 4
 
3 5 6
is the symmetric equation of the line.

Example
Find the equation of the line through the point A (1, 2, 3) and B (4, 4, 4), and find the co-
ordinates of the point where the line meets the plane z = 0.
Solution:
 4  1   3 
AB   4    2    2  is a vector parallel to the line passing through points A and B.
 4   3  1 
     
Let R (x, y, z) be any point on line AB, then

B (4, 4, 4)

R(x, y, z) A (1, 2, 3)

Line

OR  OA  AR.
So
OR  OA  t AB, where t is a scalar, thus
 x   1  3
     
 y    2  t  2
 z   3  1
     
OR
 x  4  3   4    3  3t   1   3 
             
y
  = OR = OB + BR  OB + (1-t) BA  4
   1  t  
 2   4     2  2t    2   t 2 
z  4  1   4    1  t   3   1 
             
The line meets z = 0 where the z-coordinate is 0. i.e. where 3 + t =0  t = -3.
Setting t = -3 in the vector equation, we obtain point
 x  1  3   8 
       
 y    2   3  2    4 
 z   3 1  0 
       
46
i.e. the line meets plane z = 0 at point  8,  4,0 

Example10.3
x2 y 4 z 7
1. a) Write the line   in form r  a  tu
3 5 2
b) Show that the line passes through (8, 14, 11).
c) Find the unit vector parallel to this line
Solution:
x2 y4 z 7
(a) t     x  2  3t , y  4  5t , z  7  2t
3 5 2
 x   2  3t   2   3 
       
 y    4  5t    4   t  5 
 z   7  2t   7   2 
       
(b) Showing that it passes through (8, 14, 11)
 x   2  3t  8
       2  3t  8 
y
    4  5 t   14    4  5t  14   t  2
 z   7  2t  11   7  2t  11 
   
Thus
 8   2  3
     
 14    4   2  5 
 11   7   2
     
 3
 
d) The vector parallel to this line is a~   5  . The unit vector in the direction of a~ is
 2
 
given by
 3 
 
 3  3  38

1
a
1   1  
5
   5
    5 
a ~
32  52  22  2 38  2  38 
~     2 
 
 38 

Exercise
 x  2  4   x   10    2
           
2. Show that the equations  y    3   m  6  and  y    15   n   3  represent the
z 1   2  z    3  1 
           
same line.

47
Planes in 3-dimensions spaces
A plane in 3-space can be specified by giving a non-zero vector that is perpendicular to the
plane called a normal and one of the points.

Suppose we want to find the equation of the plane passing through the point P0  x0 , y0 , z0 
and having non-zero vector n   a, b, c  as a nomal.
~

Let P  x, y, z  be any point on the plane;

P(x, y, z)


Then P0 P is orthogonal to n , so that
~


n . P0 P  0 1)
~

Since

P0 P   x  x0 , y  y0 , z  z0  ,

equation 1) becomes

a  x  x0   b  y  y0   c  z  z0   0 2)

We call this the point normal form of the equation of the plane.

Example 10.4

48
Find an equation of the plane passing through point (3, -1, 7) and perpendicular to the vector
n   4, 2,  5 .
~

Solution:

Let  x, y, z  be any point on the plane. Then

 x  3, y  1, z  7 
is a vector on the plane and must be perpendicular to n   4, 2,  5 . Therefore,
~

 4, 2, 5 .  x  3, y  1, z  7   0
 4  x  3  2  y  1  5  z  7   0

 4 x  12  2 y  2  5 z  35  0

Or

 4 x  2 y  5 z  25  0

or

4 x  2 y  5 z  25 .

Note: 4, 2,  5 are the coefficients of x, y, z and they are the components of the vector
perpendicular to the plane (normal to the plane).

Theorem

If a, b, c and d are constants and a, b, c are not all zero, then the graph of the equation

ax  by  cz  d

is a plane having the vector n   a, b, c  as a normal.

~

Example 10.5

Find the equation of the plane passing through the points

P1 1, 2,  1 , P2  2, 3, 1 , P3  3,  1, 2  .

Solution:

Two vectors on the plane are;

49
 
P1 P 2   2, 3, 1  1, 2,  1  1, 1, 2 


P1 P 3   3,  1, 2   1, 2,  1   2,  3, 3

A vector perpendicular to the plane is

i j k
~ ~ ~
  1 2 1 2 1 1
1 2  PP
PP 1 3  1 1 2 i j k  9 i  j 5 k .
~ 3 3 ~ 2 3 ~ 2 3 ~ ~ ~
2 3 3

Therefore by the previous theorem the equation of the plane is

9 x  y  5z  d ,

where d is a constant to determined.

Since P1 1, 2,  1 lies on the plane, it satisfies the above equation, so

d  9(1)  2  5  1  9  2  5  16 .

 equation of the plane is 9 x  y  5 z  16 .

Example 9.5

Find the equation of the plane through  1, 2, 3 and perpendicular to the planes
2 x  3 y  4 z  1 and 3x  5 y  2 z  3 .

Solution:

A vector perpendicular to the plane 2 x  3 y  4 z  1 is  2, 3, 4 

A vector perpendicular to the plane 3x  5 y  2 z  3 is  3, 5, 2 

A vector parallel to the two planes is

i j k
~ ~
3 4 2 3
~
2 4
 2, 3, 4    3, 5, 2   2 3 4  i j k  14 i  8 j  k
~ 5 2 ~ 3 2 ~ 3 5 ~ ~ ~
3 5 2

and this vector is perpendicular to the plane in question.

50
Therefore equation of the plane perpendicular to the planes 2 x  3 y  4 z  1 and
3x  5 y  2 z  3 is

14 x  8 y  z  d .

Now since the plane passes through point  1, 2, 3 ,

14  1  8  2    3  14  16  3  1 .

Therefore equation of the plane is

14 x  8 y  z  1 .

Example 10.6:

Find the angle between the following two planes 2 x  y  2 z  1 and x  2 y  2 z  2 .

Solution:

A vector perpendicular to the plane 2 x  y  2 z  1 is  2,1, 2 

A vector perpendicular to the plane x  2 y  2 z  2 is 1, 2, 2 

Angle between the two planes is the same as the angle between their normal;

cos  
 2,1, 2   1, 2, 2 
2,1, 2 1, 2, 2

224 4
 
9 9 9

4
   cos 1    63.40 .
9

51
10.2.2.2 Distance from the origin to the plane ax  by  cz  d

We know that  a, b, c  is a vector perpendicular to the plane. Place the tail of  a, b, c  at the
origin and multiply  a, b, c  with an appropriate scalar  such that

  a, b, c     a, b, c 

lies on the plane.

Since   a, b,  c  lies on the plane, it must satisfy the equation of the plane.

That is,

a   a   b  b   c   c   d .

 a 2  b2  c 2  d

d
  .
a  b2  c2
2

Now, the distance from  0, 0, 0  to the plane is

D   a ,  b,  c  a    b     c  .
2 2 2

  2 a 2   2b 2   2 c 2

  2  a 2  b2  c2 

 a 2  b2  c2

d
 a 2  b2  c2
a b c
2 2 2

d
 .
a  b2  c2
2

Example 10.7

Find the distance from the origin to the plane 2 x  3 y  z  2 .

Solution:

52
 d 2 2
D  
a b c 22  32   1
2 2 2 2
14

Exercise

1. Show that the distance between a point P0  x0 , y0 , z0  and the plane ax  by  cz  d is

given by

ax0  by0  cz0  d

D .
a 2  b2  c 2

Hence find the distance from between point 1, 4, 3 and the plane 2 x  3 y  6 z  1
.

2. Find the equation of the plane through  4,1, 3 and  2, 1, 3 and perpendicular to
the plane 2 x  3 y  4 z  5 .

Intersecting and parallel planes

Given two planes, either they intersect, in which case we can ask for their line intersection or
they are parallel in which case we can for the distance between them. Two planes are parallel
if their normals are parallel vectors.

Example
1. Find the parametric equations of the line of intersection for the planes 3x  2 y  4 z  6  0
and x  3 y  2 z  4  0
Solution:
The line of intersection consists of all points x, y, z satisfying the linear system;
3x  2 y  4 z  6 or 3 x  2 y  4 z  6 or 3 x  2 y  4 z  6
x  3y  2z  4 11 y  2 z  6 11 y  2 z  6  z is a free variable

6 2t 26 16
Now, let z  t , then y    and x   t
11 11 11 11
Therefore, the parametric équations of the line of intersection are,
26 16 6 2t
x  t y  , z t    t   
11 11 , 11 11

53
 Note : Vector equation is
 26   16 
 x   11   11 
   6   2 
 y    11  t  11 .
z 
   0   1 
   

2. The planes x  2 y  2 z  3 and 2 x  4 y  4 z  7 are parallel since their normal 1, 2, 2 

and  2, 4, 4  are parallel vectors. Find the distance between the two planes.

Solution:

To find the distance D between the planes, we may select an arbitrary point in one of the
planes and compute its distance to the other. By setting y  z  0 in the equation
x  2 y  2 z  3 , we obtain point P0  3, 0, 0  in this plane. The distance between P0 and the
plane 2 x  4 y  4 z  7 is

2  3  4  0    4  0   7 1 1
D   .
22  42   4  36 6
2

Exercise
 x  1 1
     
1. Show that the line  y    2   t 1 lies on the plane 2x +3y-5z = -7.
 z   3 1
     

 3 1 1  2
       
2. Show that the lines r   5   m  2  and s   2   n  3  do not meet i.e. they are skew
7 1  3 5
       
lines.
3. Find the equation of the plane through  1, 2, 4  and containing the line of intersection

of the planes 5 x  y  z  1 and x  6 y  z  2 .

Assessment Questions
1. Find the equation of the plane passing through the point 3,1,7 and perpendicular to the
vector n  4,2,5 .

54
2. Find the line through the point 1, 2, 3 and parallel to the vector v   4, 5, 7  .

3. (a) Find parametric equations for the line l passing through the points P1 2,4,1 and
P2 5,0,7 .

(b)Where does the line intersect the xy -plane?

5. Find parametric equations for the line of intersection of the planes 3x  2 y  4 z  6  0 and
x  3 y  2 z  4  0.

55
 LINEAR DEPENDENCE AND INDEPENDENCE OF VECTORS
Linear combinations
Definition (Linear Combination)

A vector w is called a linear combination of the vectors v1 , v2 , , vk provided there exists

~ ~ ~ ~

scalars c1 , c2 , , ck such that

w  c1v1  c2 v2   ck vk .
~ ~ ~ ~

Example 11.1

Determine whether vector w   2, 6,3 3 is a linear combination of the vectors

~

v1  1, 2, 1 and v2   3, 5, 4  .

Solution:

Set w  c1v1  c2 v2 . That is

~ ~ ~

 2, 6,3  c1 1, 2, 1  c2  3, 5, 4  .

Equating the corresponding components we obtain the linear system,

c1  3c2  2
2c1  5c2  6 .
c1  4c2  3

Reducing the augmented matrix to echelon form;

 1 3 2  R2  2 R1  1 3 2  1 3 2 
    R3  7 R2  
 2 5 6    0 1 2    0 1 2  .
 1 4 3  R  R  0 7 5   0 0 19 
  3 1    

We see from the third row that the system is inconsistent, so the desired scalars c1 and c2 do
not exist. Hence w is not is a linear combination v1 and v2 .
~

Example 11.2

56
Express w   7, 7, 11 as a linear combination of the vectors
~

v1  1, 2, 1 , v2   4,  1, 2  and v3   3, 1, 3  .

~ ~ ~

Solution:

Set c1v1  c2 v2  c3 v3  w , that is

~ ~ ~ ~

c1 1, 2,1  c2  4, 1, 2   c3  3, 1, 3   7, 7, 11 .

Thus

c1  4c2  3c3  7
2c1  c2  c3  7
c1  2c2  3c3  11

We solve this linear system;

 1 4 3 7  R2  2 R1  1 4 3 7   1 4 3 7   1 4 3 7 
    1 7 R2   R3  6 R2  
 2 1 1 7    0 7 7 21  0 1 1 3   0 1 1 3  .
 1 2 3 11  R  R  0 6 6 18    0 6 6 18  0 0 0 0 
  3 1      

Thus

c1  4c2  3c3  7
infinitely many solutions
c2  c3  3 

Thus, w can be written as a linear combination of v1 , v2 and v3 in infinitely many ways.

~

Now, c3 is a free variable. Set c3  t , then c2  3  t and c1  5  t . For instance, t  1 gives

c1  4, c2  2 and c3  1 , so

w  4 v1  2 v2  v3 .
~ ~ ~ ~

But t  2 yields c1  7, c2  5 and c3  2 , so w can also be expressed as

~

w  7 v1  5 v2  2 v3
~ ~ ~ ~

and so on.

57
11.2.2 Linear Dependence and independence
Definition (Linear dependence): If v1 , v2 , , vk are vectors in  n , we say that v1 , v2 , , vk
~ ~ ~ ~ ~ ~
are linearly dependent if one vector can be written as linear combination of the others.

Eg.
 7, 7,11 ,  4, 1, 2  , 1, 2,1 and  3,1,3 in the previous example are linearly dependent.
Theorem
Vectors v1 , v2 , , vk   n are linearly dependent if and only if there exists scalars
~ ~ ~

c1 , c2 , , ck  not all zeros such that

c1v1  c2 v2   ck vk  0 .
~ ~ ~ ~

Proof:
 Suppose that v1 , v2 , , vk are linearly dependent, so one of the vectors say vk is a linear
~ ~ ~

combination of v , v , ,v ;
~1 ~ 2 ~ k 1

v  c1 v  c2 v   ck 1 v ,
~k ~1 ~2 ~ k 1

so

c1 v  c2 v   ck 1 v  v  0
~1 ~2 ~ k 1 ~k ~

Therefore,

c1 v  c2 v   ck 1 v  ck v  0,
~1 ~2 ~ k 1 ~k ~

where ck  1  0. Thus not all ci ' s are zero.

Assume that c1 v  c2 v   ck 1 v  ck v  0, where not all ci ' s are zero, say ck  0 .

~1 ~2 ~ k 1 ~k ~

Then

ck v  c1 v  c2 v   ck 1 v .
~k ~1 ~2 ~ k 1

 c   c   c 
 v    1  v   2  v    k 1  v i.e. v is a linear combination of its
~k
 ck  ~ 1  ck  ~ 2  ck  ~ k 1 ~k

predecessors and therefore v1 , v2 , , vk are linearly dependent.

~ ~ ~

Example 11.3
Are the vectors v  1, 2, 2, 1 , v   2, 3, 4, 1 and v   3, 8, 7, 5  in  4 linearly dependent?
~1 ~2 ~3

58
Solution:
Set c1v1  c2 v2  c3 v3  0
~ ~ ~ ~

i.e.
c1 1, 2, 2, 1  c2  2, 3, 4, 1  c3  3, 8, 7, 5    0, 0, 0, 0 
Thus
c1  2c2  3c3  0
2c1  3c2  8c3  0
2c1  4c2  7c3  0, homogeneous linear system
c1  c2  5c3  0
We solve by Gaussian elimination method,

1 2 3 0  R2  2 R1  1 2 3 0 1 2 3 0
     
2 3 8 0  0 1 2 0  R4  R2  0 1 2 0
2 4 7 0  R3  2 R1  0 0 1 0  0 0 1 0 .
     
1 1 5 0  R4  R1  0 1 2 0  0 0 0 0 
Thus

c1  2c2  3c3  0
 c2  2c3  0  c1  0, c2  0, c3  0
c3  0
Since there is no free variable, the system does not have a non-trivial solution. Infact,
c1  0, c2  0, c3  0 . Hence the vectors are not linearly dependent.

Definition (Linear independence): The vectors v1 , v2 , , vk are linearly independent if they

~ ~ ~

are not linearly dependent. Hence v1 , v2 , , vk are linearly independent if

~ ~ ~

c1v1  c2 v2   ck vk  0 implies c1  c2   ck  0 .
~ ~ ~ ~

Example 11.4
Determine whether the vectors v  1,  2, 1 , v   2, 1,  1 and v   7,  4, 1 are linearly
~1 ~2 ~3
independent.

Solution:
Set
c1 1, 2,1  c2  2,1, 1  c3  7, 4,1  0   0, 0, 0  .
~
Then

59
 c1  2c2  7c3  0
2c1  c2  4c3  0
c1  c2  c3  0
Reducing the augmented matrix to echelon form,

 1 2 7 0  R2  2 R1  1 2 7 0  1 2 7 0 1 2 7 0
    5R3  3R2   15  
 2 1 4 0    0 5 10 0    0 5 10 0  R2  0 1 2 0  .
 1 1 1 0  R  R  0 3 6 0  0 0 0 0  0 0 0 0
  3 1      
Thus

c1  2c2  7c3  0  infinitely many solutions

c2  2c3  0

Since c3 is a free, the homogeneous system has a non-trivial solution. Thus the three vectors
are not linearly independent, they are linearly dependent.

Example 11.5
Show that the vectors v   6, 2, 3, 4  , v   0, 5,  3, 1 , v   0, 0, 7,  2  are linearly
~1 ~2 ~3
independent.
Solution:
Set
c1  6, 2, 3, 4   c2  0, 5,  3, 1  c3  0, 0, 7,  2   0   0, 0, 0, 0  .
~
Then
6c1 =0
2c1  5c2 0  c1  0, c2  0, c3  0
3c1  3c2  7c3  0
4c1  c2  2c3  0
Accordingly, v , v and v are linearly independent.
~1 ~ 2 ~3

NOTE:
Observe that vectors in the preceding example form a matrix in echelon form;
6 2 3 4 
 
 0 5 3 1 
 0 0 7 2 
 
Thus, we have shown that the non-zero rows of the above echelon matrix are independent;
the results holds true in general.

Theorem
60
The non-zero rows of a matrix in echelon form are linearly independent.
Example 11.6
Determine whether the following vectors in  4 are linearly dependent or independent;
i) 1, 3,  1, 4 , 3, 8,  5, 7  ,  2, 9, 4, 23
ii) 1,  2, 4, 1 ,  2, 1, 0,  3 , 3,  6, 1, 4 
Solution:
i) Form a matrix whose rows are the given vectors and row reduce to echelon form;

 1 3 1 4  R2  3R1  1 3 1 4   1 3 1 4 
    R3  3R2  
 3 8 5 7    0 1 2 5    0 1 2 5  .
 2 9 4 23  R  2 R  0 3 6 15  0 0 0 0 
  3 1   
Since the echelon matrix has a zero row, the vectors are linearly dependent.

 1 2 4 1  R2  2 R1  1 2 4 1 
ii)  2 1 0 3    0 5 8 5  .
 3 6 1 4  R  3R  0 0 11 1 
  3 1 
Since the echelon matrix has no zero row, the vectors are linearly independent.

Further Example
Write the polynomial v  t 2  4t  3 as a linear combination of the polynomials
e1  t 2  2t  5, e2  2t 2  3t , and e3  t  3 .

Solution

Set v as a linear combination of ei using the unknowns x, y and z ; v  xe1  ye2  ze3
   
t 2  4t  3  x t 2  2t  5  y 2t 2  3t  z  t  3 .
 xt 2  2 xt  5 x  2 yt 2  3 yt  zt  3z
  x  2 y  t 2   2 x  3 y  z  t  5x  3z
Equating corresponding coefficients, we obtain

x  2y 1
2 x  3 y  z  4
5x  3 z  3
Reducing the augmented matrix to echelon form;
 1 2 0 1  R2  2 R1  1 2 0 1  1 2 0 1 
    R3  10 R2  
 2 3 1 4    0 1 1 6     0 1 1 6
 5 0 3 3  R  5 R  0 10 3 8  
  3 1   0 0 13 52 
61
Therefore

x  2y 1
yz 6
13z  52  z  4, y  2, x  3

So
v = -3e1 + 2e2 + 4e3

Exercise
Determine whether or not the following vectors in  3 are linearly dependent:

a) 1,2,1, 2,1,1, 7,4,1 c) 1,2,3, 1,3,2, 2,1,5

b) 1,3,7, 2,0,6, 3,1,1, 2,4,5 d) 2,3,7, 0,0,0, 3,1,4

62
 VECTOR SPACES
Vector space
Definition (Vector space): A vector space is a non-empty V of objects, called vectors, on
which are defined two operations, called addition and multiplication by scalars, subject to the
axioms below. For all u, v, w V and all scalars c and d ;
~ ~ ~

1. The sum of u and v denoted by u  v is in V. Closure under addition

~ ~ ~ ~

2. u  v  v u . Additive commutativity
~ ~ ~ ~

3.  u  v   w  u   v w .
~ ~ ~ ~ ~ ~
Additive Associativity
4. There is a vector in V , denoted by 0 and called the zero vector, for which u  0  u for
any vector u V . Additive identity
5. For each u  V , there is a vector  u V such that u   u  0 .
~ ~ ~
 
~
Additive inverse
6. The scalar multiple of u and c , denoted c u is in V.
~ ~
Closure under scalar multiplication
 
7. c u  v  c u  c v .
~ ~ ~ ~
Left distributive law

8.  c  d  u~  c u~  d u~ . Right distributive law

 
9. c d u   cd  u .
~ ~
Scalar multiplication associativity

10. c  d u   d  c u  Scalar pull through

~ ~

11. 1. u  u . Multiplicative identity

~ ~

Example
1. The set  
n
 a , a ,
1 2 
, an  an   is a vector space.
2. M n , m   , the set of all n  m matrices over  is a vector space with respect to the
operations of usual addition of matrices and multiplication of matrices by a scalar.
3. pn  x  , the set of all polynomials a0  a1 x  a2 x2   an xn with real coefficients is a
vector space with respect to addition of polynomials and multiplication by a constant.
The zero vector in pn  x  is the polynomial 0  0 x  0 x 2   0 x n
4. Let F    ,be the set of all functions f :    , then F    is a vector space. Given two
functions f and g in F    and a real number c , the functions f  g and cf are
defined as follows;
 f  g  x   f  x   g  x 
and
 cf  x   c f  x 

63
Subspace of a subspace
Definition(Subspace): Let W be a non-empty subset of a vector space V . Then W is a
subspace of V if W is itself a vector space with respect to the operations of addition and scalar
multiplication as defined in in V.

Theorem:
A non-empty subset W of a vector space V is a subspace of V if and only if
i)  u, w W , u  w W -Closure under addition:
~ ~ ~ ~

ii)  k , u W , k u  W -Closure under scalar multiplication.

~ ~
i.e. W is a subspace of a vector space V iff its closed under vector addition and scalar
multiplication.

Examples:
1. Show that W   x, y  x  0 is a subspace of  2 .
Solution:
W is the y-axis
i) Let u   x1 , y1  , w   x2 , y2  W , then x1  x2  0 .
~ ~
Now,

u  w   x1 , y1    x2 , y2    x1  x2 , y1  y2 
~ ~

and
x1  x2  0 .

 u  w W
~ ~

ii) If k is a scalar, then

k u  k  x1 , y1    kx1 , ky1 
~

and
kx1  k.0  0  k u W .
~

Hence W is a subspace of 3 .

2. Show that W   x, y  
x 2  y 2 is not a subspace of  2 .
Solution:
x2  y 2  x2  y 2  0
  x  y  x  y   0

Either
x  y or x   y i.e. W is a union of two straight lines
We need to give a counter example;
64
Now, u  1, 1 , w   1, 1 W
~ ~

but
u  w  1, 1   1, 1   0, 2   W since 0 2  2 2 .
~ ~

Since the first axiom is not satisfied, then W is not a subspace of  2 .

3. Show that W   x, y, z  : x  y  z  0 is a subspace of 3 .
Solution:
W is a plane through the origin.
i) Let u   x1 , y1 , z1  , w   x2 , y2 , z2  W , then x1  y1  z1  0 and x2  y2  z2  0 .
~ ~
Now,

u  w   x1 , y1 , z1    x2 , y2 , z2    x1  x2 , y1  y2 , z1  z2 
~ ~

and
 x1  x2    y1  y2    z1  z2    x1  y1  z1    x2  y2  z2   0  0  0 .
 u  w W
~ ~

ii) If k is a scalar, then

k u  k  x1 , y1 , z1    kx1 , ky1 , kz1 
~

and
kx1  ky1  kz1  k  x1  y1  z1   k .0  0
.
 k u W
~

Hence W is a subspace of 3 .
Remark 12.1
1. To show W is a subspace, prove in general and to show that W is not a subspace, give
a counter example.


2. Subspaces of  2 are: 0 ,  2 , lines through the origin
~

3. Subspaces of 3 are: 0 ,  , lines and planes through the origin

3
~

Further Examples
Determine whether
a) W  ( x, y) : y  x 2  is a subspace of  2 .

b) W  (a, b, c) : ab  0 is a subspace of 3 .

65
Solution
a) W  ( x, y) : y  x 2  is not a subspace of  2 for let

u  (1,1), w  (1,1) W .
~ ~

But
u  w  (1,1)  (1,1)  (0, 2) W , since 22  0 .
~ ~

b) W  (a, b, c) : ab  0 is not a subspace of 3 for let

u  (1, 0,1), w  (01,1) W .

~ ~

But
u  w  (1, 0,1)  (0,1,1)  (1,1, 2) W , since 1.1  0 .
~ ~

Theorem
Let u1 , u2 , , um  n and W be the set of all linear combinations of u1 , u2 , , um . Show that
~ ~ ~ ~ ~ ~

W is a subspace of  n .
Proof
i) Let x, y  W , then
~ ~

x  k1u1  k2 u2   km um and y  c1u1  c2 u2   cm um .

~ ~ ~ ~ ~ ~ ~ ~

Now,
x  y  (k1  c1 )u1  (k2  c2 ) u2   (km  cm ) um W .
~ ~ ~ ~ ~

ii) If k is a scalar,
 
k x  k  k1u1  k2 u2   km um   x (k k1 )u1  (kk 2 ) u2   (kk m ) um  W .
~  ~ ~ ~  ~ ~ ~ ~

Hence W is a subspace of  n .

Exercise
1. Show whether or not the following are subspaces of 3 ;
a) W  (a, b, c) : a  b  c  1
b) W  ( x, y, z ) : y  2 x
2. Suppose U and V are subspaces of  n . Show that U V is a subspace of  n .

66
 Spanning Sets
Definition (Spanning Sets)

If u1 , u2 , , um are vectors in a vector space V, then the set of all linear combinations of
~ ~ ~

u1 , u2 , , um form a subspace of V. This subspace is said to be spanned by V.

~ ~ ~

Now, u1 , u2 , , um is called the spanning set of the subspace.

~ ~ ~

Example
1.  is spanned by (1, 0, 0), (0,1, 0), (0, 0,1) since if  x, y, z  3 , then
3

 x, y, z   x(1, 0, 0)  y(0,1, 0)  z(0, 0,1) .

2. (1,1), (2,1) span  2 since if  x, y  2 , then
a (1, 1)  b(2, 1)  ( x, y )
 ab  x b x y
ab  y a  y  ( x  y)  2 y  x

So,
( x, y )  (2 y  x)(1, 1)  ( x  y )(2, 1)

Hence (1,1), (2,1) span  2

3. Show that (1,1,1), (0,1,1), (0,1,  1) span 3 .

Solution
Let  x, y, z  3 and set
( x, y, z )  a(1, 1, 1)  b(0, 1, 1)  c(0, 1,  1)
a x  abc  y  abc  y
abc  y or a  b  c  z or  2c  z
abc  z a x ax

ax
yz
c
2
2 x  y  z
b
2
Therefore,

 2 x  y  z   yz
( x, y, z )  x(1,1,1)    (0,1,1)    (0,1, 1)
 2   2 

Since the system is consistent, then (1,1,1), (0,1,1), (0,1,  1) span 3 .
67
 4. Is (1,1) a spanning set for  2 ?
Solution:

(1,1) is not a spanning set for  2 since for example (1, 2) 2 but

Basis and Dimension

Definition (Basis)

Let B be a finite subset of a vector space V. Then B is said to be a basis for V if;

i) B spans V.
ii) B is a linearly independent set.

Examples

1. (1, 0), (0, 1) is a basis for  2 since for any  x, y  2 ,
( x, y )  x(1, 0)  y (0, 1) ,
Thus (1, 0), (0, 1) spans  2 .
Also (1, 0), (0, 1) is linearly independent since
a(1, 0)  b(0, 1)  (0, 0)  a  0 and b  0 ,
so (1, 0), (0, 1) is a linearly independent set.
Therefore, (1, 0), (0, 1) forms a basis for  2

2. Already, we have shown that (1,1), (2,1) span  2 . It can be checked that
(1,1), (2,1) is a linearly independent set. Hence
(1,1), (2,1) forms a basis for  2 .
3. (1, 0, 0), (0,1, 0), (0, 0,1) forms a basis for 3 .
4. (1, 0, 0), (0,1, 0), (0, 0,1),(1,1,1) span 3 for if  x, y, z  3
 x, y, z   x(1, 0, 0)  y(0,1, 0)  z(0, 0,1)  0(1,1,1)
But (1, 0, 0), (0,1, 0), (0, 0,1),(1,1,1) is not linearly independent since

(1, 1, 1)  1(1, 0, 0)  1(0, 1, 0)  1(0, 0, 1) ,

and so (1, 0, 0), (0,1, 0), (0, 0,1),(1,1,1) does not form a basis for 3 .

5. (1,1) is not a basis for  2 since (1,1) does not span  2 .

Remark
A basis for  n contains exactly n vectors.

68
Example
Determine whether or not (1,1,1), (1, 2, 3),(2, 1,1) form a basis for 3 .
Solution
Since the vectors are 3 in number, we only need to check for linear independence;
 1 1 1  1 1 1   1 1 1 
     
 1 2 3   0 1 2    0 1 2 
 2 1 1   0 3 1  0 0 5 
     

The echelon matrix has no zero rows, hence the three vectors are linearly independent and so
form a basis for 3 .

Definition (Dimension):

Let V be a vector space and B be a basis for V, then the dimension of V, which is denoted by
dimV is the number of vectors in B.

EG: dim 2  2, dim 3  3

Theorem

Suppose B1 and B2 are both basis of V. Then B1 and B2 have the same number of vectors,
so dimension of a vector space is non-ambiguous. It does not depend on the basis you choose.

Corollary

Dimension of  n is n.

Remark:

Let

e  (1, 0, 0, , 0)
~1

e  (0,1, 0, , 0)
~2

e  (0, 0, 0, ,1)
~n

where e has 1 in the ith position and zeros elsewhere, then e

~i
  is a basis of  . This is
~i
n

called the standard or usual basis for  n .

EG:

a) (1, 0), (0, 1) is the standard basis for  2

69
 b) (1, 0, 0), (0,1, 0), (0, 0,1) is the standard basis for 3
c) (1, 0, 0, 0), (0,1, 0,0), (0, 0,1,0),(0, 0, 0,1) is the standard basis for  4
Example

Let W be the subspace of  4 generated/spanned by vectors;

(1,  2, 5,  3), (2, 3, 1, 4), (3, 8,  3, 5) .

i) Find a basis and dimension of W.

ii) Extend the basis of W to the basis of the whole space  4 .

Solution:

i) We check for linear independence;

 1 2 5 3   1 2 5 3   1 2 5 3 
     
 2 3 1 4    0 7 9 2    0 7 9 2 
 3 8 3 5   0 14 18 4   0 0 0 0 
     

The non-zero rows (1,  2, 5,  3) and (0, 7,  9, 2) of the echelon matrix form a basis for W.
Thus in particular dimension of W is 2.

ii) We seek four independent vectors which include the above two vectors. The vectors

Alternatively;

Definition (Dimension)

The dimension of a vector space can also be defined as the number of degrees of freedom
when choosing a vector in a vector space.

It is also the number of free or independent variables when choosing a vector in a vector
space.

Examples

1. 3   x, y, z  : x, y, z 

We have 3 independent choices, hence dim  3   3 . A basis for 3 is

(2,1, 4), (0,1, 3), (0, 0, 2) .

2. V   x, y, z  : z  x  y
70
 We have 2 free variables hence dimV  2 . A basis for V is (1, 1, 2), (0, 1, 1) .
3. U   x, y, z  : x  2 y and z  3 y

We have only one free variable, hence dimU  1. A basis for U is  2,1,3
4. Find the dimension and a basis for
W   x, y, z, w : x  y  z  w  0 .
Solution:
The equation has 3 free variables i.e. y, z and w and so dimW  3
Set i) y  1, z  0, w  0
ii) y  0, z  1, w  0
iii) y  0, z  0, w  1
to obtain the respective basis
 1,1,0,0 ,  1,0,1,0  ,  1,0,0,1 .
5. Find the dimension and a basis for
W   a, b, c, d  : b  2c  d  0 .
Solution
b  2c  d  0 or 0. a  b  2c  d  0
We have 3 independent variables, hence dimW  3 .
Set i) a  1, c  0, d  0
ii) a  0, c  1, d  0
iii) a  0, c  0, d  1
Thus a basis for W is
1,0,0,0 ,  0, 2,1,0 ,  0, 1,0,1 .
Definition:
Let U and V be subspaces of  n , then


U  V  u  v : u U , v V
~ ~ ~ ~

is a subspace of  n called the sum of U and V.

Theorem
dim(U  V )  dim U  dim V  dim(U  V ) .

Example
Let
U   x, y, z, w : x  y  z  w  0, x  y  z  w  0
and
71
 V   x, y, z, w : x  y  z  w  0 .

Find the dimension and a basis of

a) U
b) V
c) U V
d) U V

Solution:

i) U   x, y, z, w : x  y  z  w  0, x  y  z  w  0
We first solve the two equations simultaneously;

x yzw0 x yzw0 x yzw0

or or
x yzw0  2 y  2w  0 y w0

There are 2 independent variables; z and w hence dimU  2

Set i) z  1, w  0

ii) z  0, w  1 ,

we obtain the respective basis

 1,0,1,0 ,(0, 1,0,1) .

ii) V   x, y, z, w : x  y  z  w  0

There are 3 free variable; y, z and w hence dimV  3 .

Set i) y  1, z  0, w  0
ii) y  0, z  1, w  0
iii) y  0, z  0, w  1

Hence a basis for V is

1,1,0,0 ,(1,0,1,0),(1,0,0,0) .
iii) U V consists of all those vectors  x, y, z, w satisfying conditions defining U and
V;

72
 x y zw0 x yzw0 x yzw0
x y zw0 or  2 y  2w  0 or  2y  2w  0 or
x yzw0 2 y  2 z  2w  0  2z 0
x yzw0
y w0
z 0

The system has one free variable i.e. w, hence dim U  V   1 . A basis of this
subspace is  0, 1, 0, 1 .
iv) dim(U  V )  dim U  dim V  dim(U  V )
 2  3 1  4 .
A basis for U  V is
 1, 0,1, 0 , (0, 1, 0,1),(1,1, 0, 0), (1, 0,1, 0) .
Exercise
Let U and W be the following subspaces of  4 ;
U   a, b, c, d  : b  c  d  0
and

W   a, b, c, d  : a  b  0, c  2d 

Find the dimension and a basis of

a) U
b) W
c) U W
d) U W

73
74
75
76