MH1100/MTH112: Calculus I.

Problem Set #9.

SOLUTIONS

Problem 1: (#3.1.9 from [Stewart].)

Sketch the graph of a function which is continuous on [1, 5], and has all the
following properties:
It has an absolute maximum at 5.
It has an absolute minimum at 2.
It has a local maximum at 3.
It has local minima at 2 and at 4.
Solutions
There are, of course, very many dierent ways of doing this problem. The
following is a simple solution:
6
5
4
3
2
1
0
1
2
1

(Note that the problem did not ask us for a dierentiable solution, in which
case the above graph would not have been a solution.)
!
1

Problem 2: (#3.1.10 from [Stewart].)

Sketch the graph of a function which is continuous on [1, 5], and has both
of the following properties:
It has no local maximum or minimum.
2 and 4 are critical numbers.
Solution:
Something like the following is what is called for:
8
7
6
5
4
3
2
1
0
1
2
1

This graph shows two dierent ways that a point can be a critical number
without being a local maximum or minimum. (Also note that according to
a strict interpretation of the definition of local max/min, the boundary
points here are not local extreme values, because f needs to be defined on
both sides of the point. This isnt an important issue though.)
!

Problem 3: (#3.1.14 from [Stewart].)

(a) Sketch the graph of a function which has two local maxima, one local
minimum, and no absolute minimum.
(b) Sketch the graph of a function which has three local minima, two local
maxima, and seven critical numbers.
Solution:
(a) A suitable function is f (x) =

x2
:
x4 +1

1.5

0.5

0.5

1
5

(x + 3)2 + 1 x 3
cos x
3 < x < 3 :
(b) A suitable function is f (x) =

2
(x 3) + 1 x 3
5
4
3
2
1
0
1
2
3
15

10

10

15

(Note that we werent asked give explicit expressions for these functions, but it is kind of fun to look for elementary expressions with the
requried properties.)
!
3

Problem 4: (#3.1.23 from [Stewart].)

Sketch the graph of the function f : [2, 5) R given by the rule
f (x) = 1 + (x + 1)2 ,
and use that graph to find and classify the extreme values of f (x).
Solution
This of course is going to be obtained from the standard parabola by elementary geometric operations in the following sequence:
1. Shift the standard parabola one to the left.
2. Shift the result one up.
So we can sketch the function directly:
50
45
40
35
30
25
20
15
10
5
0
5
6

Now we can read o the extrema. They are:

A local minimum, which is also the absolute minimum of the function,
of 1 at x = 1.
Note that because 37, the least upper bound of the function, is not an actual
value of the function, the function has no absolute maximum. Also, because
x = 2 is a boundary point of the function, it is not a local maximum
(strictly speaking).
!
4

Problem 5:
Sketch the graph of the function
$
x2
1 x < 0
f (x) =
,
2
2x
0x1
and then use that graph to find and classify the extreme values of f (x).
Solution:
In the region 1 x < 0, the function is an elementary graph which we can
draw straight in.
To plot the graph in the region 0 x 1, we take the standard parabola,
then:
1. Reflect it over the x-axis.
2. Shift the result up 2.
The final graph is:
3
2.5
2
1.5
1
0.5
0
0.5
1
2

1.5

0.5

0.5

1.5

We can now read o the extrema:

The absolute maximum of the function is 2, which occurs at x = 0.
This point is also a local maximum.
!
5

Problem 6:
Use the Closed Interval Method to find the absolute maximum and the
absolute minimum of the function f (x) = x2x+1 on the interval [0, 2].
Solution:
The derivative of this function is:
%
&
d
x

f (x) =
dx x2 + 1
=
=
=

d
dx

[x] (x2 + 1) x
(x2 + 1)2
x2 + 1 2x2
(x2 + 1)2
1 x2
.
(x2 + 1)2

d
dx

'

x2 + 1

This function is dierentiable at every point of this interval, so the critical

numbers will be the points where this derivative is zero. The derivative will
be zero at x = 1 and x = 1. So the only critical number occurring inside
the given interval is x = 1. So the values we have to compare are:
0 , occurring at the left hand end, x = 0.

1
2

, occurring at the critical number x = 1.

2
5

, occurring at the right hand end, x = 2.

Thus the absolute maximum of the function is 12 , occurring at x = 1, and

the absolute minimum of the function is 0, occurring at x = 0.
!

Problem 7: (#3.1.53 from [Stewart])

Find the absolute maximum and the absolute minimum of the function
f (t) = t 4 t2 on the interval [1, 2].
Solution:
This function is dierentiable at every point of its domain except at the
right hand end of the interval, t = 2, where it has a vertical tangent. The
derivative is:
)
(
1
d '
f (t) = 1 4 t2 + t

4 t2
2 4 t2 dt
2
)
t
=
4 t2
4 t2
*
+
1
=
4 t2 t2
2
4t
4 2t2
=
.
4 t2
Thus
the only critical number of the function inside the interval (1, 2) is
t = 2. The values we must compare are:

3, occurring at the left hand end, t = 1.

2, occurring at the critical number t = 2.

0, occurring at the right hand end t = 2.
So we can now read o:

The absolute minimum of the function is 3, occurring at t = 1.

The absolutemaximum of the function is 2, occurring at the critical
number t = 2.
!

Problem 8: (#3.1.54 from [Stewart])

Find the
absolute maximum and the absolute minimum of the function
f (t) = 3 t(8 t) on the interval [0, 8].
Solution:
This function is dierentiable at every point of the interval (0, 8) (though it
has a vertical tangent at the left hand end, 0) so the critical points inside
that interval are the points where the derivative is zero.
The derivative is:

d ,
d
3
3
f (t) =
t (8 t) + t
[8 t]
dt
dt
1
1
3 (1)
=
2 (8 t) + t
3t 3.
/
1 8
t
= 2
t
t 3 .3 3 /
2
8 4t
= t 3

.
3
3
The values we must compare, then, are:
0, occurring at the left hand end t = 0.

6 3 2, occurring at the critical point t = 2.

0, occurring at the right hand end t = 8.

Thus the absolute maximum is 6 3 2, occurring at t = 2, and the absolute

minimum is zero, which occurs at both the left hand end as well as the right
hand end of the interval.
!

Problem 9:
Find the absolute maximum and the absolute minimum of the function
f (x) = x 2 cos x on the interval [, ].
Solution:
Obviously this function is dierentiable at every point, so the critical points
will be the points where the derivative is zero. The derivative is:
f (x) = 1 + 2 sin x.
This will be zero when sin x = 12 . There will be two points x in the interval
[, ] where sin x = 12 . They are x = 6 and x = 5
6 . Thus the values
of f (x) that we have to compare to find the guaranteed extreme values are:
f () = 2 1.142.

*
+
* 5 +
5
5
f 5
6 = 6 2 cos 6 = 6 + 3 0.886.

* +
* +
f 6 = 6 2 cos 6 = 6 3 2.25.
f () = 2 cos() 5.142.

Thus the absolute maximum of the function is approximately 5.142, occurring at x = , while the absolute minimum of the function is approximately
2.25, occurring at the critical point x = 6 .
!

Problem 10: (#3.1.68 from [Stewart])

Show that 5 is a critical number of the function g(x) = 2 + (x 5)3 but that
g(x) does not have a local extreme value at 5.
Solution:
The derivative of g(x) is
g (x) = 3(x 5)2 .
Thus 5 is certainly a critical point of the function, because g (5) = 0. To
show that x = 5 is not a local extreme value, we can give two explanations. The first is an intuitive explanation, which says: this function is just
obtained by translating the standard graph of f (x) = x3 by 5 units to the
right and 2 units up. The original function x3 has no extreme values, so the
translated function 2 + (x 5)3 will have no extreme values either.
A more precise explanation could use the Increasing/decreasing test,
which well prove later in this problem set. Note that the derivative
g (x) = 3(x 5)2
is positive at all points to the left of x = 5, and is also positive at all points
to the right of x = 5. So the function is increasing at all points (by the
Increasing/Decreasing test) and so cannot have any extreme values.
!

10

Problem 11: (#3.1.72 from [Stewart])

Recall that a cubic function is a function of the form
f (x) = ax3 + bx2 + cx + d
where a = 0.
(a) Give examples to illustrate that it is possible for a cubic function to
have 2, 1, or no critical numbers.
(b) How many local extreme values is it possible for a cubic function to
have?
Solution:
It suces to give examples of the three possibilities.
Case I: (No critical numbers). The example f (x) = x3 +x is a cubic function
whose derivative
f (x) = 3x2 + 1
is never zero. Because this function is a polynomial, and so dierentiable
at every point, there are no critical numbers. Here is the graph:
8
6
4
2
0
2
4
6
8
2

1.5

0.5

0.5

1.5

Case II: (One critical number). The example f (x) = x3 is a cubic

function whose derivative f (x) = 3x2 is clearly zero at exactly one point,

11

x = 0. Here is the graph:

8
6
4
2
0
2
4
6
8
2

1.5

0.5

0.5

1.5

Case III: (Two critical numbers). The example f (x) = x3 x is a cubic

function whose derivative is
f (x) = 3x2 1.
This is zero at exactly two point x =

1
3

and x = 13 . Here is the graph:

8
6
4
2
0
2
4
6
8
2

1.5

0.5

0.5

1.5

What is the situation in general? Well... the derivative of a cubic function is a quadratic function, that is, a function of the form
f (x) = Ax2 + Bx + C .
By the completing the square procedure, we know that such a function can
only have 0, 1 or 2 roots. So cubics can have at most 2 local extreme values.
Part III above shows that there are concrete examples of such functions.
!
12

Problem 12: (#3.2.11 from [Stewart])

Verify that the function f (x) = 3 x satisfies all the assumptions of the Mean
Value Theorem on the closed interval [0, 1]. Then find all the possible points
c which fit the conclusion of the theorem.
Solution:
This function is certainly continuous on the interval [0, 1]. (Recall that
we proved explicitly in the - part of the course that limxa x1/3 = a1/3 .)
Also, the function is dierentiable on the open interval (0, 1). (This
example has some interest because if the assumptions of the MVT were
slightly dierent, and required dierentiability on [0, 1] instead of (0, 1), then
this function would be disqualified because of the vertical tangent at 0.) At
any rate, this function certainly satisfies the assumptions of the MVT.
So the MVT predicts that there exists a c in (0, 1) where:

3
1 2
f (1) f (0)
10

f (c) = c 3 =
=
= 1.
3
10
1
We can solve this for the only such c, giving:

1 2
c 3 =1
3
2
1
c3 =
3
1
2
c =
27
1
c=
3 3

3
c=
.
9
!

13

Problem 13: (#3.2.16 from [Stewart])

Consider the function f (x) = 2 |2x 1|. Prove that there does not exist
a point c where
f (3) f (0) = f (c)(3 0).
Why does this not contradict the Mean Value Theorem?
Solution
To clarify the discussion, well rewrite f (x) explicitly as a piecewisedefined function. When x 12 , then |2x 1| = (2x 1), and when x 12 ,
then |2x 1| = (2x 1). So:
0
3 2x x 12
f (x) =
.
1 + 2x x 12
From this we deduce that the derivative is:

2
x>

undefined x =
f (x) =

2
x<

1
2
1
2
1
2

Our task is to show that there does not exist a point where f (3)f (0) =
f (c)(3 0). If we rearrange this equation and substitute in values we learn
that such a c would need to satisfy
f (c) =

f (3) f (0)
3 1
4
=
= .
30
3
3

Looking at our formula for the derivative we observe that f (x) never takes
this value.
This doesnt contradict the Mean Value Theorem because this function
doesnt satisfy all the assumptions of the MVT on the interval [0, 3]. It is
definitely continuous on [0, 3], so it satisfies the first assumption, but it is
not dierentiable on (0, 3), because it is not dierentiable at x = 12 . Thus it
fails the second assumption.
!

14

Problem 14:
Apply standard theorems of calculus to deduce that equation
1 + 2x + x3 + 4x5 = 0
has exactly one real root.
Solution:
To begin, we observe that the function f (x) = 1 + 2x + x3 + 4x5 has at
least one real root. (Recall that a real root means a real number x
where f (x) = 0. We say real root because functions can have more roots
over the complex numbers.) The reason it has at least one root is basically
because the leading term is an odd power of x, x5 . Alternatively, we can
use the IVT, noting that:
f (0) = 1 > 0
f (1) = 6 < 0
f (x) is continuous on [1, 0].
So the IVT says that f has a root in (1, 0).
That was the easy bit. Next we need to show that there cannot be a
second root. How to show this? Well, assume that there were two roots, x1
and x2 . It would then follow from Rolles theorem (because this polynomial,
of course, is dierentiable everywhere) that there exists a point c between
x1 and x2 where f (c) = 0. But the derivative is:
f (x) = 20x4 + 3x2 + 2
and will never be zero. So the second root is impossible.
!

15

Problem 15: (#3.3.20 from [Stewart])

Apply standard theorems of calculus to deduce that the equation
x4 + 4x + c = 0
has at most two real roots (for any choice of constant c).
Solution:
To show that this equation has at most two real roots, well assume
that there are three real roots, x1 < x2 < x3 , (i.e. f (x1 ) = 0, etc.) and well
deduce an impossibility from that, using Rolles theorem.
Well, assume that there were three such roots. Looking at the pair x1
and x2 , Rolles theorem says that there exists a point, lets call it x12 such
that x1 < x12 < x2 and such that
f (x12 ) = 0.
Furthermore, applying Rolles theorem to the pair x2 and x3 , we learn that
there will exist a point x23 such that x2 < x23 < x3 where
f (x23 ) = 0.
Lets summarize our deductions so far: if the original function has three
roots, then Rolles theorem says that the derivative
f (x) = 4x3 + 4
has at least two roots.
But that is impossible! The derivative of the derivative,
(f ) (x) = 12x2
is only non-positive at a single point, so it must be an increasing function.
It is impossible for an increasing function to have two roots.
So our original assumption, that f (x) has three roots, was impossible.
(An alternative way to complete the deduction here is to observe that
the function x3 is increasing, which means that the function 4x3 + 4 will also
be increasing, so it must be 1-1.)
!

16

Problem 16: (#3.3.24 from [Stewart])

Assume that some function satisfies 3 f (x) 5 for all values of x. Show
that
18 f (8) f (2) 30.
Solution:
This is just the Mean Value Theorem applied for this f (x) to the interval
[2, 8]. The MVT says that there will exist a c (2, 8) such that
f (c) =

f (8) f (2)
.
82

In other words, for some c (2, 8):

f (8) f (2) = 6f (c).
Now we were told that for every possible x, the derivative f (x) lay between 3 and 5. So:
3 f (c) 5

18 6f (c) 30

18 f (8) f (2) 30.

!
Problem 17: (#3.3.25 from [Stewart])
Does there exist a function f (x) such that f (0) = 1, f (2) = 4, and f (x)
2 for all x?
Solution:
No, there cannot! Why? According to the MVT there would have to exist
a c (0, 2) such that
f (c) =

f (2) f (0)
5
= > 2.
20
2

17

Problem 19: (#3.3.28 from [Stewart])

Suppose that f is an odd function which is dierentiable at every point.
Prove that for every positive number b, there exists a number c (b, b)
such that f (c) = f (b)
b .
Solution:
We apply the MVT to the interval [b, b], and deduce that there exists
a c (b, b) where
f (b) f (b)
f (c) =
.
b (b)
The function is odd, so this c has the property that:
f (c) =

f (b) f (b)
f (b) + f (b)
f (b)
=
=
.
2b
2b
b
!

Problem 20: (#3.3.29 from [Stewart])

Use the Mean Value Theorem to prove the inequality
| sin x sin y| |x y|
for all x and y.
Solution:
Assume that a < b. (Note that
| sin a sin b| < |a b| if and only if | sin b sin a| < |b a|
so this case implies the case b < a.)
Applying the MVT we learn that there exists a c such that:

This implies that:

Thus:
as was required.

sin a sin b
= cos(c).
ab
2
2
2 sin a sin b 2
2
2 = |cos c| 1.
2
2
ab
|sin a sin b| |a b|

!
19

Problem 21: (Based on #3.3.30 from [Stewart])

Use the Mean Value Theorem to prove that if a function satisfies f (x) = c
for all x, where c is some constant, then there exists a constant d such that
f (x) = cx + d.
Solution
Define a function
g(x) = f (x) cx.
By the rules for dierentiation, this function is dierentiable at every point
with derivative:
g (x) = f (x) c = c c = 0.
In the lectures we used the MVT to prove that any function whose
derivative is zero at every point is a constant function. Thus there exists
some constant d such that
g(x) = d.
Substituting g(x) = f (x) cx concludes the proof.

Problem 22: (#3.2.34 from [Stewart])

A fixed point of a function f (x) is a number a such that f (a) = a. Let f (x)
be a function which is dierentiable at every point, and which satisfies the
property that f (x) is never 1. Prove that such a function cannot have more
than one fixed point.
Solution
Assume that this function f (x) has two dierent fixed points: an a1
where f (a1 ) = a1 , and an a2 where f (a2 ) = a2 . Well show that this implies
that f (x) must have its derivative equal to 1 somewhere, which would be a
contradiction.
To do this we basically just apply the MVT to the interval [a1 , a2 ]. It
tells us that there exists a point c where
f (c) =

f (a2 ) f (a1 )
a2 a1
=
= 1.
a2 a1
a2 a1

20

Problem 23: (The increasing/decreasing test.)

Let f (x) be a function which is continuous on some closed interval [a, b],
and dierentiable on (a, b). Prove that:
(a) If f (x) > 0 for all x (a, b), then f (x) is increasing on [a, b].
(b) If f (x) < 0 for all x (a, b), then f (x) is decreasing on [a, b].
Solution
Well focus on the proof of the increasing case: Part (a). The decreasing case is exactly the same, but with a few signs changed.
Let x1 < x2 be an arbitrary pair of points from [a, b]. Our task is to
show that f (x1 ) < f (x2 ).
Note that the function f (x) satisfies the assumptions of the MVT on
[x1 , x2 ], which are that
f (x) is continuous on [x1 , x2 ].
f (x) is dierentiable on (x1 , x2 ).
The Mean Value Theorem concludes that there exists a point c (x1 , x2 )
where
f (x2 ) f (x1 )
f (c) =
.
x2 x1
Rearranging this we get:

f (x2 ) f (x1 ) = f (c) (x2 x1 ).

This is greater than zero because x1 < x2 , and because f (c) > 0 (because
c must lie in (a, b)).
Thus f (x1 ) < f (x2 ).
!

21

Problem 24 : (#11.26 from [Spivak])

Let f be a function and M > 0 a real number such that f (x) M > 0
for all x [0, 1]. Prove that there is an interval of length 14 in [0, 1] on which
|f (x)| M
4 .
Solution
Our task is to prove the existence of an interval of length 14 in [0, 1] where
M
either f (x) M
4 or f (x) 4 . (It may help to draw a picture here to
visualize this).
The first thing to observe is that because the derivative is always > 0,
this f (x) is an increasing function (by the Increasing Test proved in the
previous problem).
One approach is to split the deduction into 2 cases, depending on the
value the function takes at the point 1/4.
* +
Case 1: f 14 M
4 .
M
Because
' f 1(x)
( is an increasing function,
' 1 ( this implies that f (x) 4 for
every x 0, 4 . So the interval x 0, 4 is the required interval.
* +
Case 2: f 14 > M
4 .

* The
+ way well deal with this case is to use the MVT to deduce that
f 34 M
4 . Once we know that then, because
' 3 ( we know that f (x) is increasing, we can conclude that the interval 4 , 1 has the required properties.
* +
To deduce f 34 M
4 :
'1 3(
By the* MVT
applied
to
the
interval
4 , 4 , we know that there exists a
+
point c 14 , 34 such that
. /
. /
.
/
3
1
3 1
1

f
f
= f (c)

M .
4
4
4 4
2
*1+
So because in this case f 4 > M
4 we deduce that
. /
. /
3
1
M
M
M
M
f
f
+
>
+
=
.
4
4
2
4
2
4
!
22