Section 4.2: The Mean Value Theorem

Section 4.
2: The Mean Value Theorem

Before we continue with the problem of describing graphs using cal-
culus, we shall briefly pause to examine some interesting applications
of the derivative. In previous sections, we examined the intermediate
value theorem - a result which guaranteed that a function had to take
certain values at certain points. In this section, we consider similar
ideas for the derivative.
1. Rolles Theorem
Before we consider the mean value theorem, we consider a related re-
sult.
Result 1.1. (Rolle’s Theorem) Suppose that f (x) satisfies the follow-
ing three properties:
(i ) f is continuous on the closed interval [a, b],
(ii ) f is differentiable on the open interval (a, b),
(iii ) f (a) = f (b).
Then there is a number c in (a, b) such that f ′ (c) = 0, or equivalently,
f (x) has a horizontal tangent line for some c in (a, b).
Naively, this is completely obvious. If we draw any two points with the
same y-value in the plane as follows:
then if we try to connect these points with the graph of a continuous

and differentiable function, it will have to have a horizontal tangent line
somewhere. Specifically, if f (x) is not constant, then it will either have
to increase at some point or decrease at some point. Consequently, to
get back to the same y-value, it will have to decrease or increase later,
and thus somewhere in between it must have switched from increasing
to decreasing meaning there was a slope of 0.
1
2
We illustrate with some examples.

Example 1.2. Show that f (x) = sin (2πx) satisfies Rolle’s Theorem
on [−1, 1] and then find all numbers which satisfy the conclusion.
Clearly f (x) is differentiable and continuous on the interval [−1, 1].
Plugging in the endpoints, we have f (−1) = 0 = f (1). This means
Rolle’s Theorem applies, so there must be some c in (−1, 1) such that
f ′ (c) = 0. Specifically, we have f ′ (x) = 2π cos (2πx), and this is equal
to 0 when x = −3/4, −1/4, 1/4 and 3/4 (so there are four values of c
such that f ′ (c) = 0 on (−1, 1)).
Example 1.3. Explain using Rolle’s Theorem why f (x) = x + ex has
only one zero.
We know
lim x + ex = ∞
x→∞
and
lim x + ex = −∞,
x→−∞
so since f (x) is continuous, it must pass through the x-axis at some

point. In particular, there must be some zero of f (x) at some point.
Now suppose there are two zeros - one at x = a and one at x = b.
Then on the interval [a, b], we have f (a) = 0 = f (b), and f (x) is
differentiable and continuous on this interval. In particular, Rolle’s
Theorem applies, so there must be c in [a, b] such that f ′ (c) = 0.
However, f ′ (x) = 1 + ex > 0, so Rolle’s Theorem can’t apply and thus
f (x) cannot have a second zero.
Example 1.4. Explain why the condition “f (x) is differentiable” is
necessary in Rolle’s Theorem.
Consider the function f (x) = |x| on the interval [−1, 1]. Clearly we
have f (−1) = f (1), and so it seems that Rolle’s Theorem should ap-
ply. However, f (x) has no horizontal tangent lines, so Rolle’s Theorem
obviously does not apply. This seems like it contradicts Rolle’s The-
orem, but it does not since f (x) needs to be differentiable for Rolle’s
Theorem to apply (which it is not at x = 0).
2. The Mean Value Theorem

Rolle’s Theorem is actually a special case of a much more general result
about the values a derivative can take.
Result 2.1. (The Mean Value Theorem or MVT) Suppose that f (x)
satisfies the following two properties:
(i ) f is continuous on the closed interval [a, b],
(ii ) f is differentiable on the open interval (a, b),
3
Then there is a number c in (a, b) such that

f (b) − f (a)
f ′ (c) = ,
b−a
or equivalently, f (x) has a tangent line whose slope is equal to the slope
of the line connecting the endpoints of f (x) on (a, b) (or the average
slope over the interval).
As with the Rolle’s theorem, naively, it is completely obvious why this
is true. If we draw any two points in the plane as follows:
then if we sketch the line between the two endpoints, there will be
some tangent line between the points whose slope is equal to the slope
of that line. We illustrate.
We illustrate with some examples.

Example 2.2. Find the values which satisfy the mean value theorem
for f (x) = e−2x on the interval [0, 3].
First we note that f (x) is differentiable and continuous on [0, 3], so
MVT applies. First, the average slope is
f (3) − f (0) e−6 − 1
= .
3−0 3
Next the derivative is f ′ (x) = −2e−2x . MVT says that there is some c
in [0, 3] such that
e−6 − 1
f ′ (c) = ,
3
so we can find c by setting these equal to each other and solving.
Specifically, we have
e−6 − 1
−2e−2x = ,
3
4
1 − e−6
e−2x =
6

1 − e−6
−2x = ln
6
giving

1 1 − e−6
x = − ln ≃ 0.897.
2 6
Example 2.3. Suppose f is continuous and differentiable for all x,
f (3) = 5 and −2 > f ′ (x) 6 3 for all x. What is the smallest and
largest values f (5) could be?
Using MVT, we know
f (5) − f (3)
= f ′ (c)
5−3
for some c in [3, 5]. Since −2 6 f ′ (x) > 5, it follows that
f (5) − f (3)
= f ′ (c) 6 3
5−3
and
f (5) − f (3)
= f ′ (c) > −2.
5−3
Thus
f (5) − f (3) 6 6
or
f (5) 6 11
and
f (5) − f (3) > −4
or
f (5) > 1.
Thus we have
1 6 f (5) 6 11.
3. Applications
There are two very important consequences of the mean value theorem.
They will be important to us at the end of chapter 4 and for all of
chapter 5 (and will be relevant to those taking Calculus 2).
Result 3.1. If f ′ (x) = 0 for all x in (c, d), then f (x) = k, a constant,
on (c, d).
5
Proof. We just need to show that for a fixed a in (c, d), f (a) = f (b)
for any other number b in (c, d). To see this, we apply MVT to the
interval [a, b]. Since f is differentiable and continuous on (c, d), it will
be on [a, b] too. Thus there is a number n in (a, b) such that
f (b) − f (a)
f ′ (n) = .
b−a
Since the derivative is 0 everywhere, it follows that f ′ (n) = 0, so
f (b) − f (a)
=0
b−a
or f (a) = f (b). Thus all values of f (x) are equal for x in (c, d) and
thus f (x) = k for some constant k.

A consequence of this is the following.
Result 3.2. If f ′ (x) = g ′(x) on [a, b], then f (x) = g(x) + k for some
constant k.
Proof. We simply apply the last result to the function f − g to get the
result.

Section 4.2: The Mean Value Theorem

Uploaded by

Copyright:

Available Formats

Section 4.2: The Mean Value Theorem

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Section 4.2: The Mean Value Theorem

Uploaded by

Copyright:

Available Formats

Section 4.

2: The Mean Value Theorem

then if we try to connect these points with the graph of a continuous

We illustrate with some examples.

so since f (x) is continuous, it must pass through the x-axis at some

2. The Mean Value Theorem

Then there is a number c in (a, b) such that

We illustrate with some examples.

You might also like