Section 1.8: X: Learning Objectives
Section 1.8: X: Learning Objectives
Section 1.8: X: Learning Objectives
SECTION 1.8 : x = f y ( )
LEARNING OBJECTIVES
PART A: DISCUSSION
• Sometimes, x is treated as a function of y, and we graph equations of the form
( )
x = f y in the xy-plane. These graphs must pass the Horizontal Line Test (HLT).
(
• The ordered pairs we associate with f are now of the form output, input . )
• x and y switch roles. y, not x, is our independent variable, and x, not y, is our
dependent variable. Function values correspond to x values.
• We will see how our rules and techniques for symmetry (from Sections 1.3 and
1.7) and transformations (from Section 1.4) need to be modified.
• We will develop a gallery of basic graphs similar to the one in Section 1.3.
( ) ()
• In Section 1.9, we will see that the graphs of x = f y and y = f x are
reflections about the line y = x .
( )
(Section 1.8: x = f y ) 1.8.2
PART B : GRAPHING x = f y ( )
Graphing x = f y ( )
( )
As a set of ordered pairs, the graph of x = f y is given by:
{ ( f ( y), y) y Dom f ( )} .
( )
• The graph of x = f y is typically different from the graph of y = f x ()
in the usual xy-plane.
• When we do function evaluations and point-plotting, we treat y as the
independent (“input”) variable and x as the dependent (“output”) variable.
( )
• WARNING 1: We still order coordinates of points as x, y , but they now
(
take the form output, input . )
PART C: SQUARING FUNCTION and EVEN FUNCTIONS
( )
Let f y = y 2 . f here is the same squaring function on we have always known.
WARNING 2: The graph never goes to the left of the y-axis, because
squares of real numbers are never negative.
A function f is even ( ) ( )
f y = f y , y Dom f ( )
The graph of x = f y is ( )
symmetric about the x - axis.
(
• f is decreasing on the y-interval , 0 , so x values decrease there, and
the graph moves to the left as we move up the graph on that interval.
)
• f is increasing on the y-interval 0, , so x values increase there, and
the graph moves to the right as we move up the graph on that interval.
( )
TIP 1: We could try to solve the equation x = f y for y in terms of x.
Domain Range
WARNING 4: f
Think: y Think: x
( ) ( ) ( )
• If f y = y 2 , then, once again, Dom f = , and Range f = 0, . )
The graph in Part D demonstrates this.
( )
(Section 1.8: x = f y ) 1.8.5
( ) ( )
Let f y = y 3 . Graph x = f y .
• Observe that f is increasing on , meaning that the graph moves to the right as
we move up the graph.
( )
WARNING 5: When graphing x = f y , trace the graph from bottom-to-
top (in the direction of increasing y), not top-to-bottom. A common error is
to reflect the graph about the x-axis, which can happen if you rotate your
head clockwise and draw the shape of the graph of y = x 3 .
A function f is odd ( ) ( )
f y = f y , y Dom f ( )
The graph of x = f y is ( )
symmetric about the origin.
( )
(Section 1.8: x = f y ) 1.8.6
WARNING 6: This time, horizontal shifts are more intuitive than vertical shifts.
( )
Let G be the graph of x = f y .
Let c be a positive real number.
Horizontal Translations (“Shifts”)
( )
The graph of x = f y + c is G shifted right by c units.
• We are increasing the x-coordinates.
( )
The graph of x = f y c is G shifted left by c units.
( )
The graph of x = f y c is G shifted up by c units.
The graph of x = f ( y + c ) is G shifted down by c units.
Example 1 (Translations)
( ) ( )
Let f y = y 2 . G is the central, purple graph of x = f y below.
§
( )
(Section 1.8: x = f y ) 1.8.7
The “coordinate shift method” from Section 1.4, Part F still applies:
We will translate the parabola on the left so that its turning point (its vertex)
( ) ( )
is moved from 0, 0 to 4, 1 ; that is, it is moved 4 units to the left and
1 unit up.
( )
2
Graph of x = y 2 Graph of x + 4 = y 1 , or
x = ( y 1)
2
4
( )
When finding intercepts in the x = f y setting, there are some differences.
()
• There can be at most one x-intercept, corresponding to f 0 , if it exists.
PART H: REFLECTIONS
( )
Let G be the graph of x = f y .
( )
The graph of x = f y is G reflected about the y-axis. (WARNING 7)
The graph of x = f ( y ) is G reflected about the x-axis.
The graph of x = f ( y ) is G reflected about the origin.
Example 4 (Reflections)
( )
Let f y = ( )
y . G is the upper right, purple graph of x = f y below.
Observe that x = y y = x2 ( x 0) , so G is the right half of an
upward-opening parabola.
§
( )
(Section 1.8: x = f y ) 1.8.9
( )
WARNING 8: Just as for translations (“shifts”) for the x = f y case, the
horizontal transformations are now the more intuitive ones.
( )
The graph of x = cf y is:
a horizontally stretched version of G if c > 1
a horizontally squeezed version of G if 0 < c < 1
( )
The graph of x = f cy is:
a vertically squeezed version of G if c > 1
a vertically stretched version of G if 0 < c < 1
If c < 0 , then perform the corresponding reflection either before or after the
horizontal or vertical stretching or squeezing.
Example 5 (Stretching and Squeezing)
( ) ( ) ( )
Let f y = y 2 . Consider x = cf y on the left and x = f cy on the right.
§
( )
(Section 1.8: x = f y ) 1.8.10
• Most of the graphs below differ from those in the table in Section 1.3, Part P.
• Corresponding domains and ranges are the same as those in Section 1.3.
•• Again, they can be inferred from the graphs.
•• The domain corresponds to y-coordinates this time, the range to x.
Even; 1 Even;
x=c x=
x-axis y2 x-axis
x= y Odd;
origin
x= y Neither
Odd
x = cy + d d = 0; Odd;
( ) x= 3
c0 y origin
then,
origin
Even; Even;
x = y2 x = y 2/3
x-axis x-axis
Odd; x= y Even;
x = y3
origin x-axis
1 Odd; x = a2 y 2 Even;
x=
y origin ( a > 0) x-axis
(Section 1.9: Inverses of One-to-One Functions) 1.9.1
LEARNING OBJECTIVES
PART A: DISCUSSION
• In Section 1.8, we saw that the graphs of x = y 3 and y = 3 x were the same.
Each equation can be solved for the “other variable” to obtain the other equation.
We call the corresponding cubing and cube root functions a pair of inverse
functions. The graphs of y = x 3 and y = 3 x are reflections about the line y = x .
• Inverse functions are developed by switching inputs with outputs and thus
domains with ranges. In order for a function f to be invertible (to have an inverse
function f 1 ), f must be one-to-one so that we obtain a function after the switch.
Sometimes, the implied domain must be restricted for a function to be invertible.
• In Section 1.8, we used the Horizontal Line Test (HLT) to see if an equation or
its graph represented x as a function of y. Now, we will use the HLT to see if a
function is one-to-one and invertible.
• The additive inverse of, say, the real number 3 is 3 . That is, 3 is the inverse
of 3 with respect to addition. This is because, if the two numbers are added, the
result is the additive identity, 0.
1 1
• The multiplicative inverse (or reciprocal) of 3 is . That is, is the inverse of
3 3
3 with respect to multiplication. This is because, if the two numbers are
multiplied, the result is the multiplicative identity, 1.
A pair of inverse functions have the property that, when they are composed in
either order, the result is the identity function, which outputs its input. They are
inverses with respect to function composition. (See Footnote 1.)
If f has an inverse function (i.e., f is invertible), then this inverse is unique, and
it is denoted by f 1 (called f inverse).
1
• Also, the inverse of f is f .
1
WARNING 1: f , the inverse function of f , is not typically the multiplicative
1
inverse of f . That is, typically, f 1 . The “ 1 ” is just a superscript; it is not
f
()
1 1
an exponent. However, f x is interpreted as
()
. (See Footnote 2.)
f x
f ( f ( x )) = x , x Dom ( f ) .
1 1
1
That is, f and f “undo” each other.
(Section 1.9: Inverses of One-to-One Functions) 1.9.3
Let f be the function that converts temperature measures from the Celsius
scale to the Fahrenheit scale.
1
f then exists. It is the function that converts from Fahrenheit to Celsius.
()
• For example, f 0 = 32 and f 1
(32) = 0 , because 0 degrees
Celsius corresponds to 32 degrees Fahrenheit. That is, 0 C = 32 F .
Both measures give the freezing point of water at sea level.
( )
• Also, f 100 = 212 and f 1
( 212) = 100 , because 100
C = 212 F .
Both measures give the boiling point of water at sea level.
1
A partial table for f and f is below.
1
f f
Input Output Input Output
x f x () x f 1 x()
0 32 32 0
100 212 212 100
1
A partial arrow diagram for f and f is below.
f ( f (0)) = f (32) = 0
1 1
f 1
( f (100)) = f ( 212) = 100 1
f ( f ( 32 )) = f ( 0 ) = 32
1
f ( f ( 212 )) = f (100 ) = 212
1
§
(Section 1.9: Inverses of One-to-One Functions) 1.9.4
( )
Dom f = Range f ( ) , and
1
Dom f ( ) = Range ( f ) .
1
That is, the domain of one function is the range of the other.
()
a) Let f x = x 3 on . Find f 1
, and verify that it is the inverse of f .
b) Let g ( x ) = x 3
on 0, 2 . Find g 1 .
c) Let h ( x ) = x 3
{ }
on 2, 1, . Find h 1 .
§ Solution
The inverse of each cubing function above is a cube root function.
a) f 1
( x) = 3
( )
x on , which is Range f . Let r x = 3 x on . ()
We will verify that r is the inverse of f by verifying the Inverse
Properties for f and r.
( r f ) ( x ) = r ( f ( x )) = r ( x ) = x 3 3 3
= x , x , and
( f r ) ( x ) = f ( r ( x )) = f ( x ) = ( ) = x , x .
3
3 3
x
()
b) g 1 x = 3 x on 0, 8 , which is Range g . ( )
c) h1 ( x) = 3
{ }
x on 8, 1, 3 , which is Range h . ()
The reader can investigate further in the Exercises. §
(Section 1.9: Inverses of One-to-One Functions) 1.9.5
()
The graph of y = f x in the xy-plane
passes the Vertical Line Test (VLT).
y = ± 9 x2
Now,
() ()
Equivalently, f a = f c a = c .
That is, identical outputs imply identical inputs.
(See Footnote 3.)
()
Let f x = x 2 on . Then, f is a function, but it is not one-to-one on , as
demonstrated by the figures below:
() ()
• Also, f a = f c a 2 = c 2 a = ± c , which is not equivalent to
a = c , if c 0 .
()
If we solve the equation f x = 9 , or x 2 = 9 , we obtain two solutions for x,
namely 3 and 3 . These are the two answers to the question, “Whose square
is 9?”
()
Let g x = x 2 on the restricted domain 0, . )
()
The graph of y = g x below passes the Vertical Line Test (VLT) and also
the Horizontal Line Test (HLT).
() ( )
• If we solve the equation g x = 9 , or x 2 = 9 x 0 , we obtain a
unique solution for x, namely 3. It is the unique input that yields 9.
On the graph of g above, the only point with y-coordinate 9 has
x-coordinate 3.
()
• More generally, g x = b has a unique solution for x, the unique
( )
input that yields b, whenever b is in Range g , which is 0, . )
We can define a unique inverse function g 1 .
( ) ( )
• Let Dom g 1 = Range g , which is 0, . )
() ()
• Define g 1 b to be the unique solution to g x = b ,
for every b in Dom ( g ) . For instance, g ( 9 ) = 3 . (In Example 1,
1 1
In summary …
A function f is invertible
()
f x = b has a unique solution, given by x = f 1
( b) , b Range ( f )
f is one-to-one
()
The graph of y = f x in the xy-plane passes the Horizontal Line Test (HLT).
( )
By the “Input-Output” Properties, a, b f b, a f ( ) 1
.
( )
The point a,b lies on the graph of f
The point ( b, a ) lies on the graph of f 1
.
1
To obtain the graph of f , reflect the graph of f
about the line y = x .
Example 5 (Restricted Squaring Function; Revisiting Example 4)
()
Again, let g x = x 2 on 0, . )
Partial tables for g and g 1 can be constructed as follows:
g g 1
x g x() Point x g 1 x () Point
0 0 (0, 0) 0 0 (0, 0)
1 1 (1, 1) 1 1 (1, 1)
2 4 ( 2, 4) 4 2 ( 4, 2)
3 9 (3, 9) 9 3 (9, 3)
(Section 1.9: Inverses of One-to-One Functions) 1.9.9
()
The graph of g, or y = g x , is in blue below. It passes the HLT, meaning
that g is one-to-one.
() ()
• The graph of y = g 1 x is the graph of y = x , because g 1 x = x . It
makes sense that the corresponding square root function undoes what the
(restricted) squaring function does.
( ) )
• It is also the graph of x = g y , or x = y 2 , with y restricted to 0, .
This is consistent with the aforementioned Graphical Properties of Inverse
Functions.
( )
The graph of x = g y for a function g is the reflection of the graph
()
of y = g x about the line y = x .
• This holds even if g is not one-to-one. §
(Section 1.9: Inverses of One-to-One Functions) 1.9.10
x +5
() ( x) .
3
1
Let f x = on . Find f
7
§ Solution
• What does f do to x?
( x ) = (7x 5) .
1 3
• Therefore, f
( )
• Dom f = Range f ( ) = , and Dom ( f ) = Range ( f ) = , so no
1 1
()
Step 1: Replace f x with y.
x +5
() ( x) .
3
1
Again, let f x = on . Find f
7
§ Solution
()
Step 1: Replace f x with y.
3
x +5
y=
7
Step 2: Switch x and y.
(Remember the theme of switching inputs with outputs.)
3
y +5
x=
7
(Section 1.9: Inverses of One-to-One Functions) 1.9.12
7x = 3
y +5 ( Multiplied both sides by 7 )
7x 5 = 3
y (Subtracted 5 from both sides)
(7x 5) (Cubed both sides)
3
=y
( )
3
y = 7x 5
( x ) = (7x 5)
1 3
f
1
• f is a function, because f was one-to-one.
() ( () )
3
1
f 1 = 7 1 5
=8
Absent Examples 6 and 7, we could have also solved f x = 1 for x. § ()
(Section 1.9: Inverses of One-to-One Functions) 1.9.13
x +5
() () ( )
3
3
Again, let f x = on . Let g x = 7x 5 on .
7
1
Check that g = f .
§ Solution
(
Check that g f ) ( x ) = x , x . ( )( )
Check that f g x = x , x .
x , x ,
( g f ) ( x ) = g ( f ( x )) ( f g ) ( x ) = f ( g ( x ))
3 x + 5 = f (( 7x 5) )
3
= g
7
(7x 5)
3
3
3
+5
3 x + 5 =
= 7 5 7
7
7x 5 + 5
=
3 7
= 3 x + 5 5
7x
3
=
= 3 x 7
=x
=x
1
Therefore, g = f .§
(Section 1.9: Inverses of One-to-One Functions) 1.9.14
• Show that f x =
9
5
()
x + 32 by developing a linear model for f such
()
that f 0 = 32 and f 100 = 212 . ( )
• Show that f 1
( x ) = 95 ( x 32) in three different ways:
•• Develop a linear model for f 1
such that f 1
(32) = 0 and
f 1
( 212) = 100 .
•• Begin with f x =
9
5
()
x + 32 and apply the Conceptual
Approach used in Example 6.
•• Begin with f x =
9
5
()
x + 32 and apply the Mechanical
Approach used in Example 7.
()
f x =
9
5
x + 32
( )
9
(
f 273.15 = 273.15 + 32
5
)
( )
= 459.67 F
1
Observe that the red graph for f and the brown graph for f below are
reflections about the line y = x .
FOOTNOTES
1. Identity functions and compositions of inverse functions. There are technically different
identity functions on different domains.
(See Footnote 3 below and Section 1.1, Footnote 1.)
• Let f be an invertible function that maps from domain X to codomain Y ; i.e., f : X Y .
If f is invertible, then f is onto, meaning that the range of f is the codomain Y.
f 1 maps from Y to X ; i.e., f 1 : Y X .
( )
• Let I X be the identity function on Dom f , which is X . I X : X X .
• Let IY be the identity function on Dom f ( ) , which is Y . I
1
Y
:Y Y .
1 1
• Then, f f = I X , and f f = IY .
• If g is a function such that g f = I X , then g is a left inverse of f ; f has a left inverse
{ } { }
f is one-to-one. For example, let X = 1, 2 , Y = 10, 20, 30 , f = 1, 10 , 2, 20 , and{( ) ( )}
{( ) ( )(
g = 10, 1 , 20, 2 , 30, 2 )} . Then, g is a left inverse of f .
• If h is a function such that f h = IY , then h is a right inverse of f ; f has a right inverse
{ } { } {( ) (
f is onto. For example, let X = 1, 2, 3 , Y = 10, 20 , f = 1, 10 , 2, 20 , 3, 20 , )( )}
{( ) ( )}
and h = 10, 1 , 20, 2 . Then, h is a right inverse of f , although h is not a left inverse of f .
• If f is one-to-one and onto, then f has a unique inverse function that serves as both a
unique left inverse and a unique right inverse.
(Section 1.9: Inverses of One-to-One Functions) 1.9.16
1
2. f n . Many instructors reluctantly use the f notation to represent the inverse function of f .
( )( )
In Chapters 4 and 5, we will accept that sin 2 x = sin x sin x , which is the standard
interpretation.
• On the other hand (and this compounds the confusion), some sources use n to indicate the
number of applications of f in compositions of f with itself; the result is called an iterated
function. For example, they would let f 2 = f f , and they would use the rule:
() ( ( )) () ()
f 2 x = f f x . This is typically different from the rule f 2 x = f x f x . ()
However, our use of the notation f 1 for “ f inverse” is more consistent with this second
1
interpretation, since f f is an identity function, which could be construed as f 0 in this
context.
()
• A one-to-one (or “injective”) function has the property that, whenever f a = f c for ()
domain elements a and c, it must be true that a = c . That is, two outputs are equal the
inputs are equal. This definition will lead to the One-to-One Properties for exponential and
logarithmic functions in Chapter 3.
• An onto (or “surjective”) function has the property that its range is the entire codomain
(see Section 1.1, Footnote 1). This means that every element of the codomain is the image
(that is, function value or output) of some element of the domain.
• When we develop the inverse of a bijective function, we switch inputs with outputs.
•• The one-to-one property guarantees that the resulting function does not allow one
input to yield more than one output.
•• The onto property guarantees that the resulting function is defined for all elements of
the codomain of the original function, which is now the domain of the inverse function.
The onto property is often ignored in discussions of inverse functions. This is because we
typically force the codomain of the original function to be the range (or “image”) of the
original function in this context; this then guarantees the onto property. (Precalculus
sources typically avoid the term “codomain” in the first place.) If this is the case, then
“one-to-one function” and “one-to-one correspondence” are interchangeable, and we only
have to worry about the one-to-one property when it comes to invertibility.