LECTURE # 18

THEOREM

If f and g are two one-to-one functions, then their composition that is

gof is one-to-one.
PROOF
We are taking functions f: X Y and g: Y Z are both one-to-one
functions.
Suppose x1, x2 X such that
(gof) (x1) = (gof) (x2)
g(f(x1)) = g(f(x2)) (definition of composition)
Since g is one-to-one, therefore
f(x1) = f(x2)
And since f is one-to-one, therefore
x1 = x2
Thus, we have shown that if (gof) (x1) = (gof)(x2) then x1 = x2
Hence, gof is one-to-one. THEOREM
If f: XY and g: Y  Z are both onto functions, then gof:X  Z is onto.
PROOF
Suppose f: X Y and g: Y Z are both onto functions. We must show that
gof: X Z is onto. Let zZ
Since g:Y Z is onto, so for zZ,
there exists yY such that g(y)=z. Further, since f: X Y is onto, so for
yY, there exists xX such that f(x) = y.
Hence, there exists an element x in X such that (gof) (x) = g(f(x)) = g(y) = z
Thus, gof: X Z is onto. 1
 THEOREM

If f: W X, g:X Y, and h:Y Z are functions, then

(hog)of = ho(gof)
PROOF
The two functions are equal if they assign the same image to
each element in the domain, that is,
((hof)of)(x) = (ho(gof)) (x) for every x W
Computing
((hog)of)(x) = (hog)(f(x)) = h(g(f(x)))
and (ho(gof)) (x) = h ((gof)(x)) = h (g(f(x)))
Hence (hog)of = ho(gof)
REMARK: The composition of functions is associative.

EXERCISE

Suppose f:XY and g:YZ and both of these are one-to-one and onto.
Prove that (gof)-1 exists and that
(gof)-1 = f-1og-1
SOLUTION
Suppose f: X Y and g:Y Z are bijective
functions, then their composition gof: X Z is also bijective. Hence (gof)-1:
Z X exists.
Next, to establish (gof)-1 = f-1og-1, we show that
(f-1og-1)o(gof) = ix and (gof)o(f-1og-1) = iz
2
Now consider
(f-1og-1)o(gof) = f-1o(g-1o(gof)) (associative law for o)
= f-1o((g-1og)of) (associative law for o)
= f-1o(iyof) (g-1og = iy)
= f-1of (iyof = f)
= ix (f:XY)
Also
(gof)o(f-1og-1) = go(fo(f-1og-1)) (associative law for o)
= go((fof-1)og-1) (associative law for o)
= go(iyog-1) (fof-1 = iy)
= gog-1 (iyog-1 = g-1)
= Iz (g:YZ)
Hence f-1og-1 = (gof)-1
REAL-VALUED FUNCTIONS

Let X be any set and R be the set of real numbers. A function f:XR that
assigns to each xX a real number f(x) R is called a real-valued function.
If f: R R, then f is called a real-valued function of a real variable.

EXAMPLE
1.f: R+ R defined by f(x) = log x is a real valued function.
2. g:R R defined by g(x) = ex is a real valued function of
a real variable.

3
 OPERATIONS ON FUNCTIONS

SUM OF FUNCTIONS
Let f and g be real valued functions with
the same domain X. That is f:X R and g:X R.
The sum of f and g denoted f+g is a real valued function with the same
domain X i.e. f+g: X R defined by
(f+g)(x) = f(x) + g(x)  xX

EXAMPLE
Let f(x) = x2 + 1 and g(x) = x + 2 defines functions f and g from R to R.
Then (f+g) (x) = f(x) + g(x)
= (x2 + 1) + (x + 2)
= x2 + x + 3 xR
which defines the sum functions f+g: X R

DIFFERENCE OF FUNCTIONS

Let f: X R and g:X R be real valued functions. The difference of f and

g denoted f-g is a function from X to R defined by
(f-g)(x) = f(x) - g(x)  xX
EXAMPLE
Let f(x) = x2 + 1 and g(x)=x+2 define functions f
and g from R to R. Then (f-g) (x) = f(x) - g(x)
= (x2 + 1) - (x + 2)
= x2 - x - 1 xR 4

which defines the difference function f-g: X R

 PRODUCT OF FUNCTIONS

Let f: X R and g:X R be real valued functions. The product of f and g

denoted f.g or simply fg is a function from X to R defined by
(f . g)(x) = f(x) . g(x)  xX
EXAMPLE
Let f(x) = x2 + 1 and g(x)=x+2
define functions f and g from R to R.
Then (f . g) (x) = f(x) . g(x)
= (x2 + 1) . (x + 2)
= x3 + 2x2 + x + 2 xR
which defines the product function f . g: X R

QUOTIENT OF FUNCTIONS

Let f:XR and g: X R be real valued functions. The quotient of

f by g denoted f is a function from X to R defined by
g
 f f ( x)
  ( x)  g ( x) is not equal to 0
g g ( x)
EXAMPLE
Let f(x) = x2 + 1 and g(x) = x + 2 defines functions f and g from R to R.
Then f f ( x)
 ( x)  x  X & g ( x)  0
g g ( x)
x2 1
 x  X 5
x2 f
which defines the quotient function : X R.
g
 SCALAR MULTIPLICATION

Let f:X R be a real valued function and c is a non-zero number. Then the
scalar multiplication of f is a function cf: R R defined by
(cf)(x) = cf(x) xX
EXAMPLE
Let f(x) = x2 + 1 and g(x) = x+2 defines functions f
and g from R to R. Then
(3f - 2g)(x) =(3f)(x) - (2g)(x)
=3  f(x) - 2  g(x)
= 3(x2+1) - 2 (x+2)
= 3x2 - 2x-1 xX

EXERCISE

If f:R R and g:R  R are both one-to-one, is f+g also one-to-one?

SOLUTION
f+g is not one-to-one
As a counter example; define f:R R and g: R R by
f(x) = x and g(x) = -x xR
Then obviously both f and g are one-to-one
Now
(f+g)(x) = f(x) + g(x) = x + (-x) = 0 xR
Clearly f+g is not one-to-one because
(f+g)(1) = 0 and (f+g) (2) = 0 but 12
6
 EXERCISE
If f:RR and g:R R are both onto, is f+g also onto? Prove or give a
counter example.
SOLUTION
f+g is not onto.
As a counter example, define f:R R and g:R R by
f(x) = x and g(x) = - x xR
Then obviously both f and g are onto.
Now (f+g)(x) = f(x) + g(x)
= x + (-x)
=0 xR
Clearly f+g is not onto because only 0R has its pre-image in R and no
non-zero element of co-domain R is the image of any element of R.
EXERCISE
Let f:R R be a function and c( 0)R.
1. If f is one-to-one, is cf also one-to-one?
2. If f is onto, is cf also onto?
SOLUTION
1. Suppose f:R R is one-to-one and c( 0)R
Let (cf)(x1) = (cf)(x2) for x1, x2  R
 cf (x1) = cf (x2) (by definition of cf)
 f(x1) = f(x2) (dividing by c0)
Since f is one-to-one, this implies
x1 = x2 7

Hence cf; R R is also one-to-one.

2. Suppose f:R R is onto and (c0) R.
Let y R. We search for an x R such that
(cf) (x) = y (1)
 cf(x) = y (by definition of cf)
y
 f(x) = (dividing by c0)
c
y
Since f: R R is onto, so for R, there exists some xR
c
such that the above equation is true; and this leads back to equation (1).
Accordingly cf: R R is also onto.

EXERCISE

The real-valued function 0x:X R which is defined by

0x(x) = 0 for all x X
is called the zero function (on X).
Prove that for any function f: X R
1. f + 0x = f 2. f 0x = 0x
SOLUTION
1. Since (f + 0x)(x) = f(x) + 0x(x)
= f(x) + 0
= f(x) xX
Hence f+0x =f
8
2. Since (f  0x)(x) = f(x)  0x(x)
= f(x)  0
=0
= 0x(x) xX
Hence f  0x = 0 x

EXERCISE

Given a set S and a subset A, the characteristics function of A, denoted A,

is the function defined from S to the set {0,1} defined as

1 if x  A
A(x) = 
0 if x  A
Show that for all subsets A and B of S
1. AB = A  B
2. AB = A + B - A  B
3. A(x) = 1 -  A(x)
SOLUTION

1. Prove that AB = A  B

Let x  AB; therefore x A and x B. Then
AB(x) = 1; A(x)=1; B(x)=1
Hence AB(x) = 1 = (1) (1) =A (x) B(x)
=(A  B ) (x) 9
 SOLUTION

Next, let y (AB)

 y A  B 

 y A or y B
Now y (AB)  (AB)(y) = 0
and y Aor y B  A(y) = 0 or B(y) = 0
Thus AB(y) = 0 = A (y) B (y)
= (A  B)(y)
Hence, AB and A  B assign the same number to each element x in S, so
by definition
AB = A B

SOLUTION

2. Prove that AB = A + B - A  B

Let xAB then x A or x B
Now  AB (x) = 1 and  A (x) = 1 or  B (x) = 1
Three cases arise depending upon which of  A(x) or  B(x) is 1.
CASE-I ( A(x)) = 1 &  B(x) = 1)
Now  A (x) + B (x) -  A (x)  B (x)
= 1 + 1 - (1) (1)
=1
=  AB (x)
10
CASE-II ( A (x) = 1;  B (x) = 0)
Now  A (x) +  B (x) - A (x) B (x)
= 1 + 0 - (1) (0)
=1
=  AB (x)
CASE III ( A (x) = 0;  B (x) = 1)
Now  A (x) +  B (x) - A (x) B (x)
= 0 + 1 + (0) (1)
=1
=  AB (x)
Thus in all cases
 AB(x) = 1 =  A (x) +  B (x) - A (x) B (x) xAB
Next let yAB. Then y(AB)
 yAB (DeMorgan’s Law)
 yA and yB
 y  A and y  B
Thus  AB(y) = 0;  A (y) = 0; B (y) = 0
Consider A (y) +  B (y) - A (y) B (y)
=0+0-0
=0
=  AB(y)
Hence for all elements of S
11
 AB= A + B- A  B
3. Prove that A(x) = 1 -  A(x)
Let xA. Then x A and so
A(x) = 1 and A(x) = 0
 A(x) = 1 = 1 - 0 = 1 - A(x) (1)
Also if yA, then yA and so
A(y) = 0 and A(y) = 1
 A(y) = 0 = 1 - 1 = 1 - A(y) (2)
By (1) and (2), for all elements of S
A(x) = 1 - A(x)
EXERCISE
If F, G and H are functions from A = {1,2,3} to A what must be true if.
1. F is reflexive? 2.G is symmetric?
3. H is transitive, onto function?
SOLUTION
1. F is reflexive iff every element of A is related to itself i.e.aFa aA.
Also F is a function from A to A, so each element of A is related to a
unique (one and only one) element of A. Hence, F maps each element
of A to itself so that F is an identity function.

A
A
1 1

2 2

3 3 12
2. G is symmetric iff
if aGb then bGa a,bA
Now, in the present case.

A
A A
A
1 1
1 1
2 2
2 2
3 3
3 3

i.e. G is both one-to-one and onto (a bijective function)

3. H is transitive iff
if aHb and bHc then aHc. a,b,cA.
In our case

A A
A A
1 1 1 1

2 2 2 2

3 3 3 3

H is transitive, onto function if and only if it is an identity function.

13
 FINITE AND INFINITE SETS

FINITE SET:
A set is called finite if, and only if, it is the empty set or there is one-to-one
correspondence from {1,2,3,,n} to it, where n is a positive integer.
INFINITE SET
A non empty set that cannot be put into one-to-one correspondence with
{1,2,3,…,n}, for any positive integer n, is called infinite set.

CARDINALITY

Let A and B be any sets. A has the same cardinality as B if, and only if,
there is a one-to-one correspondence from A to B.
COUNTABLE SET
A set is countably infinite if, and only if, it has the same cardinality as the
set of positive integers Z+.
A set is called countable if, and only if, it is finite or countably infinite.
A set that is not countable is called uncountable.

EXAMPLE

The set Z of all integers is countable.

SOLUTION
We find a function from the set of positive integers Z + to Z that is one-to-
one and onto.
Define f: Z+ Z by
 n
 2 if n is an even positive integer
14
f(n) =  n-1
 if n is an odd positive integer
 2
Then f clearly maps distinct elements of Z+ to distinct integers. Moreover,
every integer m is the image of some positive integer under f. Thus f is
bijective and so the set Z of all integers is countable (countably infinite)

EXERCISE

Show that the set 2Z of all even integers is countable.

SOLUTION
Consider the function h from Z to 2Z defined as follows
h(n) = 2n for all nZ
Then clearly h is one-to-one. For if
h(n1) = h(n2) then
2n1 = 2n2 (by definition of h)
 n1 = n2
Also every even integer 2n is the image of integer n under h. Hence h is
onto as well. Thus h:Z 2Z is bijective. Since Z is countable, it follows that
2Z is countable.
IMAGE OF A SET

Let f : X Y be a function and A  X.

The image of A under f is denoted and defined as:
f(A) = {yY | y = f(x), for some x in A}
EXAMPLE
f Let f: X Y be defined by the arrow
diagram
1 .a
Let A = {1,2}and B = {2,3} then
2 b
3 c f(A)={b} and f(B) = {b,c} 15
4
 INVERSE IMAGE OF A SET

Let f: X Y be a function and C  Y.

The inverse image of C under f is denoted and defined as:
f-1(C)={x X | f(x) C}
EXAMPLE
Let f: X Y be defined by the arrow diagram.
f
X Y

1 a

b
2
c
3
d
4

Let C = {a},D = {b,c},E = {d} then f-1(C)={1,2}, f-1(D) = {3,4}, and

f-1(E) =

16
 SOME RESULTS
Let f: X Y is a function. Let A and B be
subsets of X and C and D be subsets of Y.
1. if A B then f(A)  f(B)
2. f(AB) = f(A) f(B)
3. f(AB)  f(A)  f(B)
4. f(A-B)  f(A) - f(B)
5. if C  D, then f-1(C)  f-1(D)
6. f-1(CD) = f-1(C)  f-1(D)
7. f-1(CD) = f-1(C) f-1(D)
8. f-1(C-D) = f-1 (C) - f-1 (D)

17