2.

) If demand is inelastic at a price p, then an increase in unit price causes an increase in rev-
enue. Likewise, a decrease in unit price causes a decrease in revenue.

3.) If demand is unitary at a price p, then an increase in unit price causes the revenue to stay
about the same.

Revenue from Loudspeakers, Cont’d. We note that demand is elastic when p = 300 since
E(300) = 3 > 1. On the other hand, when p = 100, we have that E(100) = 13 < 1, hence demand
is inelastic. Because demand is inelastic when p = 100, it follows by the above that an increase in
unit price causes an increase in revenue.

3.5 Higher-Order Derivatives

d � f � (x + h) − f � (x)
[f (x)] = lim .
dx h→0 h
d
Given that this limit exists, we write dx [f � (x)] = f �� (x), read “f double prime of x.” Continuing
in this fashion, we may consider third-, fourth-, and higher-order derivatives. We write f �� (x) for
the second derivative, f ��� (x) for the third derivative, f (4) (x) for the fourth derivative, and in gen-
eral, we write f (n) (x) for the nth derivative of f (x). Like before, it is convenient to use equivalent
d2 y �� d3 y ��� dn y
notation dx 2 (or y ), dx3 (or y ), and dxn (or y
(n)
) for f �� (x), f ��� (x), and f (n) (x), respectively.

Higher-Order Derivatives of a Polynomial. Compute the derivatives of all orders of the

Solution. We have that

f � (x) = 5x4 − 2x3 + 12x2 − 4x + 1;

d �
f �� (x) = [f (x)] = 20x3 − 36x2 + 24x − 4;
dx

d ��
f ��� (x) = [f (x)] = 60x2 − 72x + 24;
dx

d ���
f (4) (x) = [f (x)] = 120x − 72;
dx

d (4)
f (5) (x) = [f (x)] = 120; and
dx

f (n) (x) = 0 for all n ≥ 6. �

32
Higher-Order Derivatives of a Power Function. Compute the first, second, and third deriva-
tives of the power function g(x) = x2/3 . What is the domain Dg��� of g ��� (x)? Give an intuitive ex-
planation of why Dg� differs from Dg .
Solution. We have that
2
g � (x) = x−1/3 ;
3
� �
�� 2 1 2
g (x) = − x−4/3 = − x−4/3 ; and
3 3 9

� �
��� 2 4 8 −7/3
g (x) = − − x−7/3 = x .
9 3 27
Observe that we have Dg��� = {x : x �= 0} = (−∞, 0) ∪ (0, ∞). We note that this is the domain
of all derivatives of this power function. We note that the domain of g(x) is given by Dg = R. We
can explain the difference between Dg and Dg� by observing that as x approaches 0, the slope of
the line tangent to g(x) grows arbitrarily large, hence g � (x) does not exist at x = 0. �
Higher-Order Derivatives of a Composite Function. Compute the second derivative of the
composite function h(x) = (2x2 + 3)3/2 .
Solution. Using the General Power Rule from section 3.3, we have that
d 3
h� (x) =
h(x) = (2x2 + 3)1/2 · (4x) = 6x(2x2 + 3)1/2 .
dx 2
Using the General Power Rule and the Product Rule, we have that
d � 1
h�� (x) = [h (x)] = (6x) · (2x2 + 3)−1/2 · (4x) + 6(2x2 + 3)1/2
dx 2

= 12x2 (2x2 + 3)−1/2 + 6(2x2 + 3)1/2

12x2 + 6(2x2 + 3)
= (Get a common denominator.)
(2x2 + 3)1/2

6(4x2 + 3)
= √ . �
2x2 + 3
Recall that the derivative of a differentiable function f (x) at a point (x, f (x)) gives the slope of
the line tangent to f (x) at that point, i.e., the derivative of f (x) measures the instantaneous
rate of change of f (x) at the point (x, f (x)). Given that f � (x) is differentiable, it follows that
the second derivative of f (x) gives the slope of the line tangent to f � (x) at the point (x, f � (x)),
i.e., the second derivative of f (x) measures the instantaneous rate of change of f � (x) at the point
(x, f � (x)). Given that f (x) has derivatives of all orders, continuing in this way, it follows that the
derivative of f (n) (x) gives the slope of the line tangent to f (n) (x) at the point (x, f (n) (x)), i.e., the
derivative of f (n) (x) measures the instantaneous rate of change of f (n) (x) at the point (x, f (n) (x)).
Often, these higher-order derivatives have meaningful interpretations.

33
Magnetic Levitation Train, Cont’d. Recall that the position of the magnetic levitation train
from before is given by the quadratic function s(t) = 4t2 , where s(t) is the distance (in feet) cov-
ered by the train, and t is time (in seconds). We note that the first derivative of s(t) gives the
velocity s� (t) = v(t) of the train at time t. Furthermore, the second derivative of s(t) gives the
acceleration s�� (t) = v � (t) = a(t) of the train at time t. Computing the velocity and acceleration of
the train at time t, we have that v(t) = 8t and a(t) = 8. We interpret these results as saying that
at time t, the train is traveling at a speed of 8t feet per second in the positive direction, and the
train’s speed is increasing by a rate of 8 feet per second per second in the positive direction.

Ball Launch from a Rooftop. Consider tossing a ball straight up into the air from the roof
of a building. We will assume that the motion of the ball is given by s(t) = −16t2 + 24t + 120,
where s(t) is the height of the ball from the ground measured in feet, and t is the time from the
launch. Determine the height of the building. Compute and interpret the meaning of the velocity
and acceleration of the ball 3 seconds after launch.

Solution. We note that s(0) = 120, hence the building is 120 feet tall. We compute the veloc-
ity and acceleration by taking the first and second derivative of s(t), respectively. We have that
v(t) = −32t + 24 and a(t) = −32. We interpret these results as saying that at time t, the ball is
traveling at a speed of −32t + 24 feet per second in the positive direction, and the ball’s speed is
increasing by a rate of 32 feet per second per second in the negative direction. �
Unsurprisingly, higher order derivatives have interesting interpretations in economics, as well.
Definition. We say that the consumer price index (CPI) measures changes in the price level
of a fixed market basket of consumer goods and services purchased by households over time.
Use the function I(t) (a ≤ t ≤ b) to denote the CPI from year a to year b. Observe that the first
derivative of I at time t = c (a < c < b) yields the rate of change of I at c, given by I � (c).
I � (c)
Definition. We define the inflation rate of the economy at time t = c as the quantity I(c)
.

By definition, the second derivative of I at time t = c (a < c < b) yields the rate of change of
I � at c, given by I �� (c). Certainly, it is possible for I � (t) to be positive while I �� (t) is negative. We
interpret this as indicating that the economy is experiencing inflation since the CPI is increasing;
however, the rate at which the CPI is growing is decreasing. Briefly stated, inflation is slowing.

Inflation Rate of an Economy. Consider the CPI function I(t) = −0.2t3 + 3t2 + 100 from the
year 2008 to 2016 (0 ≤ t ≤ 9). Find the inflation rate at the beginning of 2013 (t = 6). (Round to
four decimal places, if necessary.) Prove that inflation was decreasing at that time.

Solution. We have that I � (t) = −0.6t2 + 6t, hence I � (6) = 14.4 and I(6) = 164.8. By definition, we
� (6)
have that the inflation rate is II(6) ≈ 0.0874 = 8.74%. We prove that inflation was decreasing at
the beginning of 2013 by showing that I �� (6) < 0. We have that I �� (t) = −1.2t + 6, from which it
follows that I �� (6) = −1.2. We conclude that inflation was decreasing in early 2013. �

34