CHAPTER 8

FUNCTIONS OF SEVERAL VARIABLES

• Ph.D Vuong Thi Thao Binh

• Foreign Trade University, Ho Chi Minh City Campus
• A function, f, of two variables is a rule that assigns to
each incoming pair of numbers, (x, y), a uniquely
defined outgoing number, z.

• We have used the labels x and y for the two incoming

numbers (called the independent variables) and z for
the outgoing number (called the dependent variable).

2
Z = f(M)

=f(M)

3
4
5
 Limits of Multivariable Functions
• As x approaches a, the limit of f(x) is L, written
lim f ( x) = L
x→a
if all values of are close to L for values of x that are sufficiently
close, but not equal, to the a.
• As M approaches M0, the limit of f(M) is L, written
lim f ( M ) = L
M →M 0

if all values of f(M) are close to L for all points Ms that are
sufficiently close, but do not overlap, to the M0.

6
Limits of
Multivariable Functions
• There is some similarity between
defining the limit of a function of a
single variable versus two variables.
• Z =f(x,y)=f(M): lim f ( M ) = L
M →M 0

7
 PARTIAL DERIVATIVE

• The partial derivative of f with respect to x is written as

and is found by differentiating f with respect to x, with y held

constant.
• Similarly, the partial derivative of f with respect to y is written as

and is found by differentiating f with respect to y, with x held

constant.

8
• We use curly ds in the notation to distinguish partial
differentiation of functions of several variables from
ordinary differentiation of functions of one variable.

• The alternative notation, fx, is analogous to the f′ notation

for ordinary differentiation.

9
• In general, we can consider y fixed. Then f becomes a function
of x alone, and we can calculate its derivative with respect to x.
…………………….

• Similarly, we find fy by fixing x (treating it as a constant) and

calculating the derivative with respect to y.
…………………….

10
The Cobb–Douglas Production Function
• One model of production that is frequently considered in business
and economics is the Cobb–Douglas production function:

Q(x, y) = AL K1-, for A > 0 and 0 < a < 1,

where Q is the number of units produced with L units of labor and K

units of capital. (Capital is the cost of machinery, buildings, tools,
and other supplies.)
The marginal productivity of labor 𝜕𝑄ൗ𝜕𝐿

The marginal productivity of capital 𝜕𝑄ൗ𝜕𝐾

11
 𝜕𝑄ൗ
MPL = 𝜕𝐿

Marginal product of labour is the partial derivative of

output with respect to labour.
At each point (𝐿0, K0), the marginal product of labor (MPL)
represents the estimated increase in output resulting
from a one-unit increase in labor, assuming all other
factors remain constant (ceteris paribus).
-----------------------
At the input (𝐿=125, K=64), the estimated increase in output
resulting from a one-unit increase in labor is about…., assuming
all other factors remain constant.

12
Example
• A cellular phone company has the following production
function for a smart phone:
p ( x, y ) = 50 x y
2/3 1/ 3

where p is the number of units produced with x units of labor

and y units of capital.
a) Find the number of units produced with 125 units of labor
and 64 units of capital.
b) Find the marginal productivities and state its economic
significance.
c) Evaluate the marginal productivities at and x = 125 y = 64

13
Solution p ( x , y ) = 50 x 2 / 3 1/ 3
y

a) p(x=125, y=64) = 50. 1252/3641/3 = 50.52 .4 =5000

b) MPx = 2/3 . 50 . X-1/3 y1/3 = 80/3 = 27.33

Tại mức đầu vào (x=125, y=64), nếu lao động tăng lên 1 đơn vị
thì sản lượng tăng xấp xỉ 27.33 đơn vị

14
E.g. Let production function Q = 30K2/3L1/3 (K,L>0)

• Find and state the economic significance:

Q QK' Q1, Q QL' Q2
K L
at K=27, L=64.

15
SECOND-ORDER PARTIAL DERIVATIVES

• In general, when we differentiate a function of two

variables, the thing we end up with is itself a function of
two variables. This suggests the possibility of
differentiating a second time. In fact there are four
second-order partial derivatives.

16
17
18
THE EQUALITY OF MIXED PARTIALS
Schwarz's theorem, also called Clairaut's theorem or Young's theorem.

• Looking back at the expressions obtained in the previous

example, notice that in all cases

• It can be shown that this result holds for all functions that arise
in economics. It is immaterial in which order the partial
differentiation is performed. Differentiating with respect to x
then y gives the same expression as differentiating with
respect to y then x. (In fact, there are some weird mathematical
functions for which this result is not true, although they need
not concern us.)
19
• In practice, of course, x and y may both change
simultaneously. If this is the case, then the net change
in z will be the sum of the individual changes brought
about by changes in x and y separately, so that

20
 DIFFERENTIALS

• The change in z:

• For this reason the formula is sometimes quoted with an

equality sign and written as

where the symbols dx, dy and dz are called differentials and

represent limiting values of Δx, Δy and Δz, respectively

21
• Example
If z = x3y – y3x evaluate
at the point (1, 3). Hence estimate the change in z
when x increases from 1 to 1.1 and y decreases from 3
to 2.8 simultaneously.

22
 IMPLICIT DIFFERENTIATION
• One important application of the small increments
formula is to implicit differentiation.

• Differentiate the following functions of one variable:

y = x3 + 2x2 + 5

y3 + 2xy2 - x = 5

• In the first case, y is given explicitly in terms of x,

whereas in the second case, the functional dependence
of y on x is only given implicitly.
23
IMPLICIT DIFFERENTIATION

24
25
HOMEWORK

26
27
 PARTIAL ELASTICITY
AND MARGINAL FUNCTIONS

• Elasticity of demand: Suppose that the demand, Q, for a

certain good depends on its price, P, the price of an
alternative good, PA, and the income of consumers, Y, so that
Q = f(P, PA, Y )
for some demand function, f.

• The (own) price elasticity of demand is defined to be

with PA and Y held constant.

28
 PARTIAL ELASTICITY
AND MARGINAL FUNCTIONS
• This definition of the (own) price elasticity of demand is
identical to the one given in the previous chapter, so
following the same mathematical argument presented there,
we deduce that

29
 CROSS-PRICE ELASTICITY OF DEMAND

• The cross-price elasticity of demand is defined to be

• The sign of EPA could turn out to be positive or negative depending on

the nature of the alternative good. If the alternative good is

substitutable, then Q increases as P rises, because consumers buy more
A

of the given good as it becomes relatively less expensive. If the

alternative good is complementary, then Q decreases as P rises, because
A

the bundle of goods as a whole becomes more expensive.

30
 INCOME ELASTICITY OF DEMAND

• If a good is inferior, then demand falls as income rises and E

Y

is negative. Canned vegetables, a supermarket’s own-brand

white bread and bus transportation are examples of inferior
goods. If a good is normal, then demand rises as income
rises and E is positive. Sometimes the value of E of a normal
Y Y

good might even exceed 1. These goods are called superior.

31
 EXAMPLE
• Given the demand function
Q = 100 - 2P + PA + 0.1Y
where P = 10, PA = 12 and Y = 1000, find the
(a) price elasticity of demand;
(b) cross-price elasticity of demand;
(c) income elasticity of demand.
Is the alternative good substitutable or complementary?

32
EXAMPLE
Given the demand function
Q = 500 - 3P - 2 PA + 0.01Y
where P = 20, PA = 30 and Y = 5000, find
(a) the price elasticity of demand;
(b) the cross-price elasticity of demand;
(c) the income elasticity of demand.
If income rises by 5%, calculate the corresponding
percentage change in demand.
Would this good be classified as inferior, normal or superior?

33
UTILITY

• To analyse the behaviour of consumers quantitatively, we

associate with each set of options a number, U, called utility,
which indicates the level of satisfaction.
• Suppose that there are two goods, G1 and G2, and that the
consumer buys x1 items of G1 and x2 items of G2. The variable
U is then a function of x1 and x2, which we write as
U = U(x , x )
1 2

The marginal utility of x .

i

34
MARGINAL UTILITY

• If x1 and x2 both change, then the net change in U can be

found from the small increments formula

35
Example

• Given the utility function

U = x11/4 x23/4
determine the value of the marginal utilities

when x1 = 100 and x2 = 200. Hence estimate the change in

utility when x decreases from 100 to 99 and x increases from
1 2

200 to 201.

36
THE LAW OF DIMINISHING MARGINAL UTILITY
  2U
 x 2  0
 1
 2
 U  0
 x22

The fact that this is negative means that marginal utility of

x1 decreases as x1 rises. In other words, as the
consumption of good G1 increases, each additional item of
G1 bought confers less utility than the previous item. A
similar property holds for G2.

37
E.g. An individual’s utility function is given by
U = 1000x1 + 450x2 + 5x1 x2 - 2x12 – x22
where x is the amount of leisure measured in hours per week,
1

and x is earned income measured in dollars per week.

2

Determine the value of the marginal utilities

when x = 138 and x = 500.

1 2

Hence estimate the change in U if the individual works for an

extra hour, which increases earned income by $15 per week.
Does the law of diminishing marginal utility hold for this
function?

38
INDIFFERENCE CURVES

39
 Marginal rate of commodity substitution

• The derivative, dx /dx , determines the slope of an indifference curve

2 1

when x is plotted on the horizontal axis and x is plotted on the

1 2

vertical axis.

40
41
PRODUCTION
• The marginal product of capital
• The marginal product of labour
• The marginal rate of technical substitution

42
43
44
45
OPTIMISATION

46
 UNCONSTRAINED OPTIMISATION
• Functions of one variable y = f(x)

47
For functions of two variables z = f(x, y)

• Step 1
Solve the simultaneous equations
f (x, y) = 0
x

f (x, y) = 0
y

to find the stationary points, (a, b).

Step 2
● f > 0, f > 0 and f f - f > 0 at (a, b), then the function has a
xx yy xx yy xy 2

minimum at (a, b)
● f < 0, f < 0 and f f - f > 0 at (a, b), then the function has a
xx yy xx yy xy 2

maximum at (a, b)
● f f - f < 0 at (a, b), then the function has a saddle point at
xx yy xy 2

(a, b).

48
49
• Xem

50
 CONSTRAINED OPTIMISATION

• In general, we want to optimise a function,

z = f(x, y)
called the objective function subject to a constraint
w(x, y) = M
• The method of substitution for optimizing, then use the
theory of stationary points of functions of one variable to
optimise z.

51
52
 LAGRANGE MULTIPLIERS

• To optimise an objective function

f(x, y)
subject to a constraint
w(x, y) = M
we work as follows.

This scalar is called the Lagrange multiplier. This is called the

Lagrangian function.
53
54