Composite Functions
Functions can be combined to create a new function.
e.g. Consider f (x) = x² + 2
g (x) = 2x
f g
x f (x) gf (x)
e.g. 1 3 6
2 6 12
3 11 22
DOMAIN of gf (x) * RANGE of f (x) and to continue, these
(which is equal to must lie within the DOMAIN of g (x) RANGE of gf (x)
the domain of f (x))
gf (x)
The new composite function of gf will take us straight from x to gf (x). Here is the simplified
function:
Function f acts first.
gf (x) = g [ x² + 2 ]
= 2 [ x² + 2 ]
= 2 x² + 4 Then function g acts,
doubling the previous
expression.
The order is most important, as we see below:
g f
x g (x) fg (x)
1 2 6
2 4 18
3 6 38
DOMAIN of fg (x) *RANGE of g (x) and to continue, these
(which is equal to the must lie within the DOMAIN of f (x) RANGE of fg (x)
domain of g(x) )
fg (x)
fg (x) = f [ 2x ]
= [ 2x ]² + 2
= 4 x² + 2
fg (x) means that g acts first, then f. gf (x) means that f acts first, then g.
NOTE fg (x) gf (x)
* The composite function will ONLY exist if the range of the first function lies within the domain of
the second function.
Example 1
Functions f (x) and g (x) are defined by f (x) = 2x + 3 and g (x) = x² + 5.
Find:
a) the function fg (x)
b) the function gf (x)
c) the function g² (x)
d) the values of p so that fg (p) = 45
a) fg (x) = f [ x² + 5 ]
= 2 [ x² + 5 ] + 3
= 2 x² + 10 + 3
= 2 x² + 13
b) gf (x) = g [ 2x + 3 ]
= [ 2x + 3 ]² + 5
= 4 x² + 12x + 9 + 5
= 4 x² + 12x + 14
c) g² (x) = g g (x) = g [x² + 5 ]
= [x² + 5 ]² + 5
= x4 + 10x² + 25 + 5
= x4 + 10x² + 30
d) fg (x) = 2 x² + 13 (see answer to part a) )
fg (p) = 45 felly 2 p² + 13 = 45
2 p² = 32
p² = 16
p =
p =
Example 2
Function f has a domain of ( and is defined by f (x) = 2e².
Function g has a domain of and is defined by g (x) = 3 1n x.
a) Explain why gf (-1) does not exist.
b) Find, in its simplest form, an expression for fg (x). Identify the domain and range of fg.
a) WHY gf (-1) DOES NOT EXIST:
f acts first f (-1) = 2e-1 = 0.7358 (to 4 d.p.)
The domain of the second function g (x) is and 0.7358 does not lie within this range, so
we cannot continue. This is why gf (-1) does not exist.
b) fg (x) = f [3 1n x] g acts first
= 2e[3 1n x]
The Logarithm Rule
= 2e1n x³
a 1n x = 1n xa
= 2 x³ so...3 1n x = 1n x3
The DOMAIN of fg (x) is the domain of the function which acts first, namely g (x), therefore the
domain is .
To find the RANGE of fg (x) I will sketch a graph of the composite function 2x3. As you can see, it is
a cubic function:
1 x
DOMAIN
RANGE
We can see from the graph that the RANGE is .
We include the 2 – square brackets as the 1 lies within the domain.
Example 3
The function f has a domain of and is defined by f (x) = 2x – k where k is a constant.
a) In terms of k, write the range of f.
ANSWER: f (x) = 2x – k is a linear function.
Gradient = 2
Crosses the ‘y’ axis at (0, -k)
As the gradient of the function is positive, this means that it is an increasing function and the lowest
value of f (x) will therefore be f (1)....because 1 is the lowest value of the domain.
f (1) = 2 (1) – k
= 2–k ...and then the values of f (x) will increase.
RANGE
The function g has a domain of [0, ] and is defined by g (x) = 3x2 + 4.
b) Find the highest value for k so that the function gf can be formed.
ANSWER: To form gf, the range of the function acting first (f (x)) must lie within the domain of
the second function g (x), which is . Therefore
The highest possible value of k is 2.
c) Given that gf (2) = 31 find the value of k.
ANSWER: g f (2) = 31 f acts first, replacing x = 2
g [ 2 (2) – k ] = 31
g[4–k] = 31
3 [ 4 – k ]2 + 4 = 31
Then g acts
3 [ 4 – k ]2 = 27
[ 4 – k ]2 = 9
4–k = Expanding (4-k)2 would
4–k = complicate the equation so -4
4 = k from both sides immediately in
order to solve the equation.
1 or 7 = k
BUT in part b) of the question, note that the highest possible value of k is 2 therefore k = 1.
