Arba Minch University

College of Business and Economics

Department of Economics

Econometrics II (Econ 3051)

MODULE

Prepared By: Alemayehu Acha (MSc)

December, 2022/23
 Mathematical Economics

Table of Contents
Chapter 1 ............................................................................................................................. 3
Review of Basic Concepts .................................................................................................. 4
Introduction ......................................................................................................................... 4
1.1 The concept of calculus. ............................................................................................... 4
1.1.1 Differential calculus ................................................................................................... 5
1.1.2 Integral calculus ....................................................................................................... 14
1.2 Basic matrix manipulation .......................................................................................... 21
Chapter -2.......................................................................................................................... 24
Derivative in use .............................................................................................................. 24
Introduction ....................................................................................................................... 24
2.1 Application of Derivatives .......................................................................................... 24
2.2. Elasticities as Logarithmic Derivatives ..................................................................... 26
2.3. Higher order Derivatives............................................................................................ 30
2.4 Linear Approximations ............................................................................................... 34
2.5 Polynomial Approximations ....................................................................................... 36
2.6 Expansion of Functions using Maclaurin and Taylor series ....................................... 38
2.7 Intermediate Value Theorem (mean) .......................................................................... 41
2.8 Multivariate Calculus .................................................................................................. 44
2.9 Homogenous functions & Euler’s Theorem ............................................................... 47
Chapter 3………………………………………………………………………………..50
Unconstrained Optimization ............................................................................................ 52
3.1. Functions of one variable ........................................................................................... 52
3.2. Functions of several independent Variables .............................................................. 53
Chapter 4 ........................................................................................................................... 62
Constrained Optimization ................................................................................................. 62
Introduction ....................................................................................................................... 62
4.1. One Variable constrained optimization ..................................................................... 62
4.2. Two variable problems with equality constraint ........................................................ 64
4.3. Inequality Constraints & The theorem of Kuhn and Tucker ..................................... 71
Chapter -5.......................................................................................................................... 78
Comparative static analysis.............................................................................................. 78
Introduction ....................................................................................................................... 78
5.1. Differentiation and its application to comparative static Analysis ............................ 78

1
University of Gondar Lecture Notes
 Mathematical Economics

5.2. Comparative Statics of General function model ........................................................ 81

5.3. Limitations of comparative static analysis ................................................................. 83
CHAPTER 6 ..................................................................................................................... 83
Dynamic equilibrium analysis ......................................................................................... 83
Introduction ....................................................................................................................... 84
6.1 Definition of differential equation: - ........................................................................... 85
6.1.2 First order linear differential equation ..................................................................... 86
6.2 Methods of solving differential equation .................................................................... 86
6.3 First order linear differential equation with constant coefficient and constant term .. 92
6.4 Practical Application of First Order Linear Difference Equation in Economics ........ 94
6.5 Difference Equation .................................................................................................... 99
6.5.1 Solving Difference Equation.................................................................................. 100
6.6. DYNAMICS OF FIRST ORDER LINEAR DIFFERENCE EQUATIONS ........... 103
6.7 Practical Application of Difference Equation in Economics .................................... 105
6.7.1 The Cobweb Model............................................................................................ 105

2
University of Gondar Lecture Notes
 Mathematical Economics

Introduction to the module

The course aims to develop the mathematical ability of students up to the level required
for a general economics degree course (i.e. one not specializing in mathematical
economics) or for a modular degree course in economics.
The module introduces mathematical techniques that will be new to most students
through examples of their application to economic concepts. It also tries to get students
tackling problems in economics using these techniques as soon as possible so that they
can see how useful they are. Students are not required to work through unnecessary
proofs, or wrestle with complicated special cases that they are unlikely to encounter
again.
In this module, we introduce the basic concepts of differentiation followed by examples
of economic applications. Further developments of the topic, such as the second-order
conditions for optimization, partial differentiation, and the rules for differentiation of
composite functions, are then gradually brought in over in the subsequent chapters, again
in the context of economics application.

The module tries to cover those mathematical techniques that will be relevant to students
of economics degree program. Most applications are in the field of microeconomics,
rather than macroeconomics.
Definition: Mathematical economics is an approach to economic analysis, in which the
economist makes use of Mathematical symbols in the statement of the problem and draws
upon known mathematical theorems to aid in reasoning.

Mathematical Economics Vs Econometrics

Econometrics:- deals with the study of empirical observations using statistical methods
of estimation and hypothesis testing.
Mathematical Economics:-refers to the application of mathematics to the purely
theoretical aspects of economic analysis.

3
University of Gondar Lecture Notes
 Mathematical Economics

Chapter One
Review of Basic Concepts
1.1. Aims and objectives

After completing this chapter students will be able to :

- recall the definition and geometric interpretation of derivative

- use the different rules of differentiation to find the derivative of different types of
functions
- apply the concept of derivative in economic applications
- apply the techniques they learn in solving optimization problems in economics
- Recall the antiderivative of a function.
- Explain the different rules of integration.
- Apply the concept of integral calculus in solving economic problems.
- recall the concept of matrix and basic matrix manipulation;

Introduction
This chapter is concerned with revising the basic concepts related with quantitative
methods in economics. These concepts are discussed in detail in quantitative method
for economists I&II courses. So, the discussions in this chapter will be very brief.

1.1 The concept of calculus.

Definition: calculus, in simple terms, in a mathematical technique which is used to deal
with change, motion and growth. It is a result of great advancement in the area of
mathematics which is used to solve mathematical problems that can’t be solved using
ordinary methods. In economics, it is frequently used to solve static and dynamic
economic problem. Differential calculus, which indicates the rate of change of the
dependent variable with respect to the independent variable, is used in static economic
analysis; On the other hand, Integral calculus, difference equation & differential equation
are used in dealing with dynamic economic problems. These applications are dealt with
in a separate chapter but for this section you will be reminded about only the basic
concepts which you have dealt with in your Quantitative methods course.
4
University of Gondar Lecture Notes
 Mathematical Economics

1.1.1 Differential calculus

1. Derivatives
Given a function y  f x , the derivative of f x  at any point ‘x’ is given by
f  x  x   f  x 
 f  x  : if this limit exists then f x  is said to be
dy
 im
dx x 0 x
differentiable function at ‘x’.
f  x  x   f  x 
The term , is simply slope of the functions which is an equivalent
x
y
expression to , and it represents average rate of change of a function .But the
x
derivative approximates the rate of change of the positive function when x is
infinitesimal i.e. slope of a function at a point or instantaneous rate of change.
Example: find the derivative of f ( x)  x 2

Solution: f x  x  x  x  f x  h  subsituting.x  h

2

f x   x 2

df  x  f x  h   f x 
2
 f  x   im
dx h 0 h
( x  2 xh  h )  x
2 2 2
 lim
h
 im 2 x  h  2 x
h 0

Differentiability and continuity

If the derivative of a function, say f x , exists at point ‘c’ then f x  is said to be
differentiable at ‘c’.
To say a function f x  is differentiable at ‘c’ the following conditions should be
fulfilled
1. f x  should be continuous at ‘c’

2. f x  should have a unique (only one ) slope at ‘c’

The 1st condition requires the fulfillment of continuity conditions i.e.
1. im f x  & im f x  should exist
x c x c

5
University of Gondar Lecture Notes
 Mathematical Economics

2. im f x   im f x   im f x 

x c x c x c

3. im f x   f c 
x c

From this we can understand that continuity doesn’t necessarily mean

differentiability.
Continuality rules out jumps and gaps at a point while differentiability condition
requires smoothness of the curve at the given point in addition to continuity
conditions. In other words, differentiability rules out jumps, gaps, sharps & cusp
corners, while continuity rules out only the first two
Example show the function f x   x  2 is not differentiable at x = 2.

Steps 1 Testing continuity of the function

1. im x  2  0
x 2

2. im x  2  0
x 2

3. im f x   f 2  0., so, f x  is continuous at x  2

x 2

Step 2 in this step we will see whether the limit of the difference quotient exists or
not.
Since f x  is an absolute value function, it can be redefined as follows:

 x  2, for.x 
f x   
2  x, for.x  2
To see the existence of the limit of the difference quotient, we find its left & right side
limits .i.e.
f x  h   f x  x  h   2   x  2  1
im  im
h 0 h h 0 h

f x  h   f x  2  x  h   2  x   1
im  im
h 0  h h  0 h
From the above result, we can see that the left & right side limits are not equal. There
fore the limit of the difference quotient doesn’t exist which in turn means that the
function f ( x)  x  2 is not differentiable at x=2.

6
University of Gondar Lecture Notes
 Mathematical Economics

Rules of differentiation
1) Constant function rule
Given f x   k , where K is a constant,

f x   0
2) Power function rule
f x   x 2

f  x   2 x 21  2 x

f ( x)  x n
f ' ( x )  nx n 1

Rules of differentiation involving two or more functions of the same variables

3) Sum/ difference Rule
Given f (x ) and g (x)
d
 f ( x)  g ( x)   f ( x)  g ( x)
dx
4) Product Rule :
If f (x ) & g (x) are differentiable; then
d  f ( x).g ( x)
 f ( x) g ( x)  f ( x) g ( x)
dx
Corollary:- Given  ( x)  f ( x).g ( x), h( x)

 (x )  f ' x .g x hx   f x g ' x hx   f x g x h ' x 

Note that the above corollary is a direct extension of the product rule of
differentiation. The rule follows the same pattern for any number of functions and
it can be proved using the same rule.

7
University of Gondar Lecture Notes
 Mathematical Economics

Example: Given AR = 25 –Q2, find MR?

Dear distance learners, I think you know from preceding economics courses
about the relation ship between the marginal, average and total revenue functions.
If you have forgotten the relation ships please revise your micro economics
module. The solution for the above problem can be found by using the relation
ship between the marginal, average and total revenue functions
TR dTR dAR.Q
AR   AR.Q , MR   using the product rule of
Q dQ dQ
differentiation, we get
dAR dQ
MR  Q  AR
dQ dQ



d 25  Q 2 Q  
 25  Q 2 
dQ
 2Q 2  25  Q 2
 25  3Q 2

5) Quotient Rule
Given f (x ) and g (x) , where g x   0

d  f ( x) 
 g ( x) f ( x) g ( x)  f ( x) g ( x)
dx g ( x)2
ax 2  b
Eg. Given f ( x)  , find f (x) ?
cx
Since f (x ) is a quotient of two functions, let g ( x)  ax 2  b and h( x)  cx .
g x  g ' h  h' g
 f ( x)   f ' x  
hx  h2
g '  x   2ax, h'  x   c
2ax.cx  c ax 2  b 
 f ' x  
cx 2
ax 2  b

cx 2

8
University of Gondar Lecture Notes
 Mathematical Economics

Rules of Differentiation Involving functions of Different Variables

6) Chain Rule ( composite function rule)
Given Z  f ( y ), and y  g (x)
dz dz dy
 .
dx dy dx

Eg. Given Z  x 2  3 x  2  , find.

10 dz
?
dx
Solution: let y  x 2  3x  2 then Z   y 
10

dz dz dy dy
  10 y 9
dx dy dx dx
 10 y 9 2 x  3
 
 10 x 2  3 x  2 2 x  3
7) Inverse function Rule
Given y  f (x)
dx 1

dy dy
dx
y 1 dy
Eg. Given x  , find ?
y 1 dx
dx  y  1   y  1 2
 
dy  y  12
 y  12
dy 1
Using inverse function rule, 
dx dx
dy

  y  12 
= 
 2 
Example : find the derivatives of the following functions.
i) f ( x)  3x 3  4 x 2  3x  2 Using rules (1), (2) and 3 above
Solution
f x   33x 31  42x 21  31x11  0

 9 x 2  8x  3

9
University of Gondar Lecture Notes
 Mathematical Economics

ii) 2 x  1 Here the function is a composite function where the outer function is
6

power function.
g ( x)  2 x  1

f ( x)  x 6

f g x   2 x  1
6

df  g  x 
 f  g  x .g  x   62 x  1 .2  122 x  1
5 5

dx

iii) g  x  
3x  1
 g  x  
x 2  3x 3x  1'3x  1x 2  3x '
x 2  3x x 2  3x 2
3x 2  3x   3x  12 x  3

x 2
 3x 
2

 3x 2  2 x  3

x 2
 3x  2

Derivatives of Exponential and Logarithmic Functions

Before dealing with derivatives of exponential and logarithmic functions, it is important
to remind you about some special limits which you have discussed in your Quantitative
Methods I.
 h 1
1. im  n
h 0 h
eh 1
2. im  ne  1 : We will use these special limits in proving the derivatives
h 0 h
of exponential and logarithmic functions.
Derivatives of Exponential functions
de x
1.  ex
dx
de x e xh  e x
Proof:- by definition :-  im
dx h0 h

 im

e x eh 1 
 factoring out ex
h 0 h

10
University of Gondar Lecture Notes
 Mathematical Economics

eh 1
 im e x . im  Product rule of limits (Quant –I)
h 0 h 0 h
 e x .ne
 ex
da x
2.  a x na
dx

Proof
da x
 im
a xh  a x
 im a x
a h  1  ima x im a h  1  a x na
dx h0 h h 0 h h 0 h 0 h
da f  x 
3.  f  x .a f  x na  a f  x  f x na  Using composite function rule
dx

Derivatives of Logarithmic functions

dnx 
1.  nx  1
dx
 n x  h   nx
Proof : by definition nx  im
h 0 h
 1  x  h 
 im n    Using logarithmic properties

h 0 h
 x 
1  h
 im n1  
h 0 h
 x
1 x  h x
 im . n1    Multiplying by  1
h 0 h x
 x x
x
1  h h
 im n1  
h 0 x
 x
x
1  h h
 n.im1  
x  x
1 1
 ne 
x x
dog ax 1
2. 
dx xna

11
University of Gondar Lecture Notes
 Mathematical Economics


  nx 
Proof :- og ax 
nx

so og ax   
na  na 

 nx 
  
1
nx . Because 1 is constant
 na  na na

1 1 1 1 1 1
.   .(ln x )'  .
na x xna ln a ln a x
nx  1
x
Examples
3
e x
1) Given yx , find dy
?
dx
Solution: taking the logarithm of both sides;
ln y  e  x ln x Then taking the derivative of both sides, we get
3

y'
y
3 1
x
3
 
 3x 2 e  x ln x  e  x us ing productrule

 1
y'  y  3x 2 e  x ln x  e  x 

3

x
 
3 


3  x3 


3 1
x
3 
y'  e  x .x e   3x 2 e  x ln x  e  x 

 

2) Given y  x x find dy ?  X x (1  nx)

dx
Solution

y  x x  ln y  x ln x
ln y '  x ln x '
 ln x  1  y '  y ln x  1
y'
y
 x x ln x  1
y  e x n3 x , find
2
dy
3) Given ?
dx

12
University of Gondar Lecture Notes
 Mathematical Economics

ln y  x 2 ln 3x,

ln y '  x ln 3x '   2 x ln 3x 

y'
2 x2
y x

y '  y 2 x ln 3x  x   xy2 ln 3x  1 because(3x )' 

3 1

3x x

 xe x n3 x 1  2n3x 
2

Higher order Derivatives

Let f(x) be a function which is differentiable at a point n its domain, then f(x) is also a
function in the same domain as f(x) is differentiable. The derivative of f x , if it exists,

is called the 2nd derivative of f x  provided that it is continuous and smooth. The
derivatives found through this iteration process are called higher order derivatives.
N.B 1. the 2nd ,3rd , 4th ……… etc. derivatives are collectively called higher order
derivatives.
2. higher order derivatives are symbolized along the same line as 2nd derivative. i.e.
the 2nd , 3rd & 4th derivatives are denoted by
f x , f x , f 4
x etc or
d2y d3y d4y dy
2
, 3 , 4 Respectively, where is used as an operator symbol and the super
dx dx dx dx
scripts show the number of times that the function should be differentiated with
respect to ‘x’.

Formulas for nth Derivatives of some functions

f x  f n 
x 

1. x n nn  1n  2nn  1  n!

2. x m mm  1m  2 m  n x m  n
3. e x ex
4. a x na a x 13
5. ax  b
n
University of Gondar n! Lecture Notes
 Mathematical Economics

Self test exercise 1.1

1. check the differentiability of f x   x  1, for x  2
ax  5 , for x  2 at x=2

2. Find the first order derivatives of the following functions

a). y  x 2 log3x
b). g  x   e5 x  4 

c). f x   2 x 3  15x 2  36x  4

3. Evaluate the 2nd order derivatives of the function g x   4 x  x

1.1.2 Integral calculus

In elementary algebra we come across opposite algebraic operations. The opposite of
addition is subtraction and that of division is multiplication in the same way, the opposite
of differential calculus is Integration. It is also called anti differentiation, i. e. the process
of finding the integral of a function.

dF  x 
If F(x) is a function such that  f  x , then F(x) is the anti derivative (Integral) of
dx

f(x) symbolically this is represented as  f xdx  F x +c where ‘c’ is a constant. This
type of integral is called indefinite integral because it represents a family of functions. If
we give definite value for ‘c’, it represents definite integral (to be discussed latter.). As
opposed to definite integrals, indefinite integrals do not possess a definite numerical
value.
Rules of Integration
Since integration is the reverse process of differentiation, each rule of integration is
associated with some derivative formal.
d kx  c 
1.  kdx  kx  c 
dx
k

14
University of Gondar Lecture Notes
 Mathematical Economics

x n 1
 x dx   c.n  1
n
2.
n 1
3.  kf xdx  k  f xdx
4.   f  x   g  x dx   f  x dx   f  x dx   g  x dx

5.  e x dx  e x  k

 1
6.   x dx   e  x dx  e  x  k
e 

e mx  c
7. e mx dx 
m
 1  mx
8.  a mx dx   a  k
 m ln a 
1
9.  dx  n x  k
x

Techniques of integration
Although we have seen some rules of integration, some functions are much more
complex for such functions it is difficult to know their integrals using the above rules.
There are different techniques of finding integrals of these functions. For this chapter,
we will focus on 3 of them. I.e. Integration by substitution, integration by parts and
integration partial fraction.
Integration by substitution
One of the methods (techniques) which are used to find integrals of complex
functions is substitution method or change of variables. This method consists of
transforming the integral and the variable of integration to some standard form say ‘u’
& ‘du’ completely eliminating ‘x’ & ‘dx’.

15
University of Gondar Lecture Notes
 Mathematical Economics

Illustration evaluate x 2 x  1dx  since the integrant is the product of two terms x

2 x  1 . Where x is factor of 2x+1 we can use substitution.

Let U  2x  1
du
 2  dx  du
dx 2

 substituting the value of ‘U’ & du in place of 2x+1 & dx we get.

du
x u
2
, but here we have two variables ‘x’& U. so, we should completely

eliminate ‘x’
 U  2x  1
2x  U  1
U 1
x
2

 U  1 U du
- substituting the value x 
U 1
2 2 2
1
U  1 udu Factoring out the constants
4


1  32 1
2 
 
4 
 u  u du

1 32 1

4  u du   u 2 du

1  2 52 2 32 
4  5
  u  u k
3 

1u 
5 3
u
  k
2 2

2  5 3 


1  2 x  1   2 x  1 3 2 
5

   k
2

2  5  
  3 

Note:-  mostly, the functions to be substituted by ‘u’ (important ‘u’ s) are found in
i) in Exponents
ii) under roots

16
University of Gondar Lecture Notes
 Mathematical Economics

iii) inside brackets

iv) in the denominators

Integration by parts
Some times, we may not be able to find integrals of products of functions by using
substitution rule. We use a technique if integration called integration by parts. It is
derived from the product rule of differention.

 f x.g xdx  f xGx   f xGxdx. where Gx is anti- derivative of g x . Letting
Gx   v & f x   u we get  udv  uv   vdu.

To see how it is derived from the product rule of differention, let us start from the rule
d  f  x G  x 
 f  x G  x   g  x  f  x 
dx

f x Gx    f x Gx dx   g x  f x dx  Taking the integral of both sides.

 f x.g xdx  f x.Gx   f xGxdx

 udv  uv   vdu
 xe dx
x
Illustration evaluate

For simplicity in using integration by parts we should select the one which is easy to
integrate as ‘dv’ & the one which is easy to differentiate as ‘u’/ In the above example,
exdx is easier to integrate & ‘x’ is easier to differentiate
Let u  x & dv  e x dx
du  dx v   dv   e x dx  e x
Substituting, this values in the formula, we get

 xe dx  xe   e dx
x x x

 xe x  e x  k
 e x  x  1  k

17
University of Gondar Lecture Notes
 Mathematical Economics

Integration by partial fraction

This technique is used to find integrals of rational functions which can be decomposed (
reduced). The point of partial fraction integration is that integration of a rational function
can be reduced as follows by determining the roots of the polynomial in the denominator.
dx
 x  a  n x  a  c
Illustration :- use partial fraction method to integrate
xdx
 x  1x  2
First we should decompose it as follows
x A B
 
x  1x  2 x 1 x  2
 now multiplying this equation by ( x-1) (x-2) we get
x  Ax  2  Bx  1  comparing the terms on the left and right
x  Ax  2 A  Bx  B
x   A  Bx  2 A  B
The coefficient of ‘x’ on the right is A+B & on the left 1 so
A B 1
2A  B  0 
The constant on the let is o
Solving the above equation simultaneously we get A = -1 & B=2
x 1 2
 
x  1x  2 x 1 x  2
xdx  1dx 2dx
 x  1x  2   x  1   x  2
x2
2

 n x  1  2n x  2  c  n c
x 1

18
University of Gondar Lecture Notes
 Mathematical Economics

DEFINITE INTEGRALS
1 Meaning of Definite Integrals
In the preceeding sections, we have discussed that the integral of a function is a
family of functions due to the fact that functions having different constants have
the same derivative.
In this section we will revise another form of integral called the definite integral.
Let f (x ) is a continuous function and if F (x ) is the antiderivative of f (x ) then

 f ( x)dx  F ( x)  c
If we choose two values of x in the domain, say x  a and, x  b and substitute these
values in to the right side of the equation and taking the difference we get

[ F (b)  c]  [ F (a )  c]  F (b)  F (a )
This value is a specific numerical value free of the variable x as well as the arbitrary
constant c . It is called the definite integral of f (x ) from a to b .
Notation:
b
 a
f ( x ) dx - read as the definite integral of f from a to b

The numbers a and b are called limits of integration.

Letter a is the lower limit of integration.
Letter b is the upper limit of the integration.
b
Therefore, a
f ( x)dx  F ( x)]ba  F (b)  F (a )

Let us see some examples on how to find the definite integral of a function.
2
Example 1: Evaluate 0
x 2 dx

Solution: First find the antiderivative of x 2

1 3
x dx  x c
2

3
Then evaluate the indefinite integral at the upper and lower limits and take the difference
to get

19
University of Gondar Lecture Notes
 Mathematical Economics

2 1 3 2 1 3 1 3 8 8
0
x 2 dx 
3
x ] 0  ( 2)  ( 0)   0 
3 3 3 3
Definite integrals of marginal cost functions
This concept of the definite integral has several applications in economics. To evaluate
TVC from an MC function for a given value of output one simply substitutes the given
quantity into the TVC function. This is the same as evaluating the definite integral
between zero and the given quantity. For example, assume that you wished to find the
value of TVC when q = 8 and you are given the function MC = 7.5 + 0.3q2. This value
would be


8 8
TVC   MC   7.5  0.3Q 2  7.5Q  0.1Q 3
8
0  111.2
0 0

Therefore, TVC is equal to the area under the MC schedule between zero and the given
level of output.
Another application of definite integrals is to find consumers and producer’s surplus.
Recall from your Micro-economics course that, consumer surplus is the difference
between what the consumers are willing to pay and what they actually paid.
Geometrically, this is defined as the area below a demand schedule but above the ruling
price. This difference is often used as a measure of welfare.
For this part, you are advised to revisit your quantitative method for economists I module
Self Test Exercise 1.2
1. Evaluate the following indefinite integrals. (5 pts each)
3 x  11
a). x 2
 x6
dx b). x 2 x 1dx

2. find the definite integral of the following functions

 20  4 x 
3
a.
1

8
b.  0
MC where MC  7.5  0.3Q3 ,

c. What does the definite integral of MC represent? If the total fixed cost
is given by TFC=300, find the total cost function.

20
University of Gondar Lecture Notes
 Mathematical Economics

1.2 Basic matrix manipulation

A matrix is defined as an array of numbers (algebraic symbols) set out in rows &
columns. Usually Matrices are denoted by a capital letter in bold type.

The algebra of matrices is a suitable mathematics method which enable s us to handle

large systems of simultaneous equations. It can be used to
1. write large systems of equation in a compact way
2. Test the existence of solutions by using determinates
3. find the solutions ( if exists)
each of the numbers ( elements) in a matrix have their own definite position. If a matrix
has m rows & n columns, the order of the matrix become mxn read ‘m’ by ‘n’
a11 a12 a13 a1n
a a 22 a 23 a2n
A  21

a m1 am2 a3 a mn

 a11 , a12 ect are called the elements of the matrix

 The number of rows and columns of a matrix together define the dimension of a
matrix. If m=n then the matrix is called square matrix
If a matrix contains only one column (row ) it is called column(row) matrix
Matrix operation
Since matrix contain a all of numbers ( a whole block of numbers) the families algebraic
operations which are used for single numbers are not adoptively applicable, and there is a
need for new set of operation rules.

1. equality of matrices
A=B only if aij=bij
Addition & subtraction of matrices
     
A  B  aij  bij  aij  bij For each elemement bij corresponding to a ij

i.e. addition and subtraction of matrics A & B is defined as addition(subtraction) of each

pair of corresponding elements

21
University of Gondar Lecture Notes
 Mathematical Economics

Note equality, addition & subtraction of matrics is applicable only if the two matrics
have equal dimension
Example
4 9 2 0 4  2 9  0 6 9
2 1  0 7  2  0 1  7   2 8
       
Scalar multiplication
Multiplying a matrix ‘A’ by a number (scalar) is simply multiplying each element of the
   
matrix by the given scalar i.e. A   aij  aij

Example: The numbers of units of a product sold by a retailer for the last 2 weeks are
shown in matrix A below, where the columns represent weeks and the rows correspond to
the two different shop units that sold them.
12 30 
A =  
 8 15 
If each item sells for £4, derive a matrix for total sales revenue for this retailer for these
two shop units over this two-week period.
Solution
Total revenue is calculated by multiplying each element in matrix of sales quantities A by
the scalar value 4, the price that each unit is sold at. Thus total revenue can be
represented (in £) by the matrix
 4 x12 4 x30  48 120
R = 4A =   =  
 4 x8 4 x15   32 60 
Multiplication of matrices
Given two matrices Amxn .and, B pxq , the conformability condition for multiplication AB is

that the column dimension of A should be equal to the row dimension of B, i.e. AB will
be defined only if n=p. That is, in order to multiply two matrices, the first matrix must
have as many columns as the second matrix has rows. Matrices that fulfill this condition
are said to be conformable for multiplication. if the product AB is defined it will have the
dimension mxq.
Matrices A x B = AB
Orders mxn nx p = mxp

22
University of Gondar Lecture Notes
 Mathematical Economics

For example, if A is a 2 x 3 matrix and B is a 3 x 5 matrix, and then the order of the new
matrix AB will be 2 x 5. Note how ever, that BA is not defined.

The rule of matrix multiplication: If one matrix is multiplied by another matrix, the
basic rule is to multiply elements along the rows of the first matrix by the corresponding
elements down the columns of the second matrix.
 b11 b12 
 a11 a12 a13     c11 c12 
If   x  b21 b22  =   , then
 a 21 a 22 a 23  b   c 21 c 22 
 31 b32 
2x3 3x2 2x2
C11 = a11 (b11) + a12 (b21) + a13 (b31)
C12 = a11 (b12) + a12 (b22) + a13 (b32)
C21 = a21 (b11) + a22 (b21) + a23 (b31)
C22= a21 (b12) + a22 (b22) + a23 (b32)
Self test exercise 1.3
2 4 6  1 2 0 
    1 3 
1. Let M =  7 8 9  , N =  0 1 3  , and L =  
  5 2 0  0 0 1  4 5
   
Then find the following if they are conformable? (2 points each)
i) M+N iii) MXN
ii) M+L iv) N+M

23
University of Gondar Lecture Notes
 Mathematical Economics

Chapter -2
2. Derivatives in use
Aims and objectives
After completing this chapter students should be able to:
 Find the maximum or minimum point of a single variable function by
differentiation and checking first-order and second-order conditions.
 Use calculus to solve optimization problems in economics
 Derive marginal revenue and marginal cost functions using differentiation and
relate them to the slopes of the corresponding total revenue and cost functions.
 Calculate point elasticity for non-linear demand functions and relate it with
revenue function.
 Use linear and polynomial approximations to solve complex functions.
 Use Euler’s theorem to identify the properties of multivariable functions in
economics

Introduction
st
In the 1 chapter of this module, we have been discussing about derivatives and rules of
differentiation. This chapter on the other hand, focuses on the application of derivatives
with a special emphasis on their use in economics. Derivatives are widely used in static
economic analysis. This chapter focuses on some typical applications of derivatives.
However; it may not exhaustively address all of the applications.

2.1 Application of Derivatives

1. First order Derivatives:- the sign of first order derivative in important to
determine the behavior of functions in a given interval, i.e. whether a function is
increasing or decreasing can be determined using the sign of its first derivative.
Definition:-consider a function f(x) which is continuous in an interval I a, b where
b>a, f(x) is said to be
 increasing if f b  f a .or. f a   f b

24
University of Gondar Lecture Notes
 Mathematical Economics

 decreasing if f a   f b.or. f b  f a 

suppose that f x  is differentiable on the given Interval,

1. If f x   0 for all x in I, then f x  is increasing on I

2. If f x   0 for all x in I, then f x  is decreasing on I
3. More generally, the conclusion in (1) an (2) still hold if in each finite interval say
MTI there are only finitely many points, at which the assumptions of
f x   0 and f x   0 are not satisfied.

Illustrative Example

Given TR=20q-q2 find the intervals for which TR functions is

a) increasing
b) decreasing
dTR
MR   20  2q
dQ
20  2q  0 20  2q  0
20  2q 20  2q
10  q q  10
q  10; i.e TR is increasing for q<10, and it is decreasing for g>10.

Application of derivatives in Economics

1. The concept of marginal analysis:- marginal analysis is very important in dealing
with most economic functions. Mathematically, marginal functions ar the 1st
derivatives of total functions.
dTR
 MR
dQ

dTC
 MC
dQ
d
 M …..etc.
dQ
Derivative can also be used to determine the relationship between marginal & average
Economics functions. Consider the cost function.
TR  C Q 

25
University of Gondar Lecture Notes
 Mathematical Economics

dTC
Q  TC
TC dAC dQ
AC    (using the quotient rule).
Q dQ Q2
dTC 1MC  AC 
MC  - from this we can observe that , will be increasing when
dQ Q

MC>AC .
For TR function
PQ
TR = PQ & AR = P
Q
d PQ  dQ Qdp
MR   p.  (Using Quotient rule)
dQ dQ dQ

MR  p  Qdp
dQ

The equation for a) above is a representative relation between AR & MR if the market is
perfectly competitive price will be constant &
dp
 0 & MR  p  AR
dQ
dp
If the market is monopoly,  0, so MR<AR.
dQ
In general, MR can never be greater than AR. At extreme MR=AR when the market is
perfectly competitive.

2.2. Elasticities as Logarithmic Derivatives

i) Elasticity Function
Given any function y  f (x)
Let x change from x to x+h, then the proportionate change in x is given by
x  x  h   x h
 
x x x : similarly, the proportionate change in y is given by

y f  x  h   f ( x)
 .
y f ( x)

26
University of Gondar Lecture Notes
 Mathematical Economics

The ratio of the proportionate change in y per unit proportionate change in x

y f ( x  h)  f ( x )
y f ( x)
is given by: 
x h
x x
f ( x  h)  f ( x ) x
.
f ( x) h

 f ( x  h)  f ( x )  x
yx   im .
 h  y

yx  dy x
. Point elasticity of y with respect o x.
dx y

ii) Elasticity as logarithmic Derivatives

Given a function, y  f (x)

Let u  ny

v  nx  x  e v
d (ny) du du dy dx
  . .
d (nx) dv dy dx dv
dy
 1 . .e v
y dx

dy
 1 . .x
y dx

dy x
 yx  .
dx y

iii) Elasticity of Demand and Marginal Revenue (MR)

Assume: TR  PQ
dP
MR  P  Q.
dQ
 1 
MR  P1  or
 ed 

27
University of Gondar Lecture Notes
 Mathematical Economics

 1 
MR  AR1  
 ed 
From this we conclude that:
1
i) MR  0, if 1   0 or  d  1  elastic demand
d
In this case TR increases as Q increases (or as p decreases)
1
ii) MR  0, if 1   0 or  d  1  inelastic demand
d
In this case TR decrease as Q increases (or as P decreases)
1
iii) MR  0, if 1   0 or  d  1  unitary elastic demand
d
In this case TR does not change as Q( or P ) changes.

AR
Example: verify  d  for the demand function
AR  MR
a p
Q , b  0, Where Q = output
 b 
P = price
Relationship between Average cost and Marginal cost functions
Given C (Q), AC (Q) and AC (Q ), MC (Q ), for Q  0; TC (Q)  AC (Q).Q

 C (Q) 
d  
d AC (Q)  Q  1
  MC (Q)  AC (Q). AC
dQ dQ Q
MC MC
- from this it follows that:
i) At the minimum point AC curve AC
The slope is zero:
 AC (Q)  MC (Q)
ii) When AC is falling, AC (Q )  0 Q

 MC (Q)  AC (Q)
iii) When AC is rising, AC (Q )  0
 MC (Q)  AC (Q)

28
University of Gondar Lecture Notes
 Mathematical Economics

Relative Vs Absolute Extremum

- consider the following three cases:

Y y y
E

A B C D

F
X x x
(a) (b) (c)

In figure (a), no significant choice regarding the value of x is made for the maximum
or minimum of y. Hence the points represent neither a maximum nor minimum.
In figure ( b) , given the set of non- negative real numbers as the domain, the function
is strictly increasing. However we may consider point D representing an absolute
(global) minimum in the range of the functions.
In figure c, points E and F are examples of a relative (or local) maximum and
minimum respectively.
Summary: Condition Relative max Relative min
1st order necessary condition f ( x0 )  0 f ( x0 )  0

2nd order sufficient condition f ( x0 )  0 f ( x0 )  0

Note that : For f n ( x0 )  0, where n is higher derivative of order n;

i) if f n ( x0 )  0 and n is even, the point is an extremum point

ii) If f n ( x0 )  0 and n is odd , the point is an inflection point

Example: 1) Given y  x 5  f ( x)

29
University of Gondar Lecture Notes
 Mathematical Economics

f ( X )  5 x 4


Y y  x5

f 5 ( X )  120  The function is at a point of

inflection, at x= 0

2.3. Higher order Derivatives

2.3.1. Concavity and Convexity
Definition: let f x  be a function which is defined on an interval I, we say that f x  is
f b   f a 
convex (concave upward) on I if f c   Lc , where L x   f a   x  a  for
ba
all a, b, c XI with a< c <b. Here, Lx  is the line through a, f a  and b, f b . we say that
it is concave down on I if f c   Lc for all a, b, c XI with a < c< b.
Geometrically: the line tangent to convex (concave upward) functions is below the graph
of the functions (tangent from below).
The tangent line of concave down ward functions is tangent from i.e. the line is above the
graph.

30
University of Gondar Lecture Notes
 Mathematical Economics

Fig.

(a) Decreasing and convex (b) decreasing and concave

(c ) increasing and convex (d) increasing and concave

For a function f x  defined on an interval I,

1) If f x  is differentiable and f x  is increasing on I, then f x  is concave

upward on I and if f x  is decreasing on I, then f x  is concave down on I.

2) If f x  is twice differentiable on I, then

f x  is concave upward on I if f x   0 for all XXI
f x  is concave down on I if f x   0. for all X X I
Illustrative Example
Given the cost function TC  0.5Q 3  0.7Q 2  30Q  256, then
a) find the values MC
b) find the interval for which AC & MC are concave upward & concave down ward.
Solution:-
dTC
MC   1.5Q 2 1.4Q  30
dQ
TC 0.5Q 3  0.7Q 2  30Q  256
AC  
Q Q
256
 0.5Q 2  0.7Q  30 
Q

d 2 AC
 AC will be concave upward if 0
dQ 2

31
University of Gondar Lecture Notes
 Mathematical Economics

dAC 256
 Q  0.7 
da Q

d 2 AC 512
2
 1 3  0
dQ Q
512
1
Q3

Q 3  512
Q  8  Concave upward (convex)
Q<8concave downward

Example :1 Given, f ( x)  ax 2  bx  c, for a>0

Given , check the concavity or convexity of the function?
 since f ( x)  2a  0 it is a strictly convex function
Examples: check for concavity and /or convexity
a) f ( x)  ax 2  bx  c Answer f ( x)  2a : - concave when 2a <0
a<0
- Convex when 2a>0
a>0
b. Given the production function y  AK  , A  0,0    1

d2y
  (  1) AK   2  0  concave.
dk 2

c. Suppose the functions U & g are both increasing and concave so that
U   0, U   0, g   0, and g   0. prove that t he composite function
f ( x)  g (U ( x )) is also increasing and concave?
Solution : f ( x)  g (u ( x))
d g (us  du( x)
f ( x)  .
du( x) dx

32
University of Gondar Lecture Notes
 Mathematical Economics

 g  (u ( x)) . u ( x)
0 0
 0 ( increaing )

 The increasing transformation of an increasing function is increasing.


f ( x)  g (u( x).u ( x)

 g (u( x)) .u ( x)  g (u( x).u ( x)

 g  u ( x)) .u ( x)  u ( x). g (u ( x) 

2

 0 0 0 0 
(-) _ (-) =  0 concave.
 An increasing concave transformation of a concave function is concave

Self test Exercise 2.1

1) Given Q  p  , where  and  are constants: find price elasticity of demand?
2) Given y  f (x ), show that the derivative


d logay  dy
,a  c  . x

d logax
 dx y ?

3) Given 1 x 3  1 x 2  2 x  1 , find the critical points and determine wether they

9 6 3
are maximum or minimum.
4) Given the total cost function:
C (Q)  aQ 2  bQ  c, (Q  0). find the minimum point of the average cost function?
5) Find the intervals where the following cubic cost function is convex and where it is
concave and find the point of inflection?
C (Q)  aQ 3  bQ 2  cQ  d

6) for the function y  ( x  8) 4

a) Find the critical value (s)?
b) Test to see if at the critical values the function is at a relative max, min or
possible inflection point?

33
University of Gondar Lecture Notes
 Mathematical Economics

2.4 Linear Approximations

To avoid working with a complicated function we sometimes try to find a simpler
function that approximates the original one.
Since linear functions are easy to work with, we first try to find a linear
approximation for any given function.
Suppose the function y=f(x) is differentiable at x= a, what is the equation of the
tangent to the graph of the function at the point (a, f( a))?

- Let us approximate the function y= f(x)

Y by the tangent line passing through the
point ( a, f(a)).
f(a) y=f(x)

a x x

- Given the equation of the tangent line as :

y  mx  b, the slope of the function

At a is given by f ( a )  m
y  f (a ).x  b
Thus, f (a )  f ' (a ).a  b

34
University of Gondar Lecture Notes
 Mathematical Economics

Thus, we have :  b  f (a)  f (a).a

At point a,
 y  f ( a ). x  f ( a )

y  f ( a )  f ( a )( x  a )

Example
1) Find the linear approximation of f ( x)  3 x about x= 1
y  f (a)  f (a)( x  a)
 f (1)  1
 2
f (1)  1 (1) 3
1
3 3

 3 x  1  1 ( x  1) for x close to 1
3

2) Find the linear approximation of f    (1  3 2   1 2  1 ) 2 about   0 ?

1

Solution f (0)  1
1
 1 
f ( )  1  3    2 
 2 2 
2
3 2   . 12
And f (0)  1 2 3 2 
3
4
 
1
 f ( )  (1  3   1  2 ) 2
 f (0)  f (0)(  0)
2 2

 1 3 
4

3) Find an approximate value for: (1.001)50?

Let f ( x)  x 50 , x  1

x 50  f (1)  f (1)( x  1)
 1  50(1.001  1

 1.05

35
University of Gondar Lecture Notes
 Mathematical Economics

2.5 Polynomial Approximations

Approximations by linear functions may be insufficiently accurate. So, it is normal to try
quadratic approximations or approximations by polynomial of higher order.
1) Quadratic Approximations:- How can a function f(x) be approximated near
x=a, by a quadratic approximation?

Given p( x )  f ( x )  A  B( x  a )  C ( x  a ) 2

p ( x)  B  2C ( x  a )
p( x )  2c
When x=a
p (Q)  A  f (a )
p (a )  B  f ( a )
1 1
p (Q )  2c  c  p (a )  f (a )
2 2
Hence pxA  Bx  a  cx  a  is approximated about x  a as:
2

f x   f a   f a x  a   1 f a x  a 

2
2

Example. Find the quadratic approximation for:

f ( x)  3 x , about x  1

Solution: f x   x 3 , and f (1)  1

1

2
f  x   1 x 3
, f (1)  1
3 3
5
f  x    2 x 3
, f (1)   21
9 9

Thus, f ( x)  1  13 x  1  19 x  1
2

2) Higher order approximations

Suppose we want to approximate a function f(x) around x=a by an nth degree
polynomial of the form:
px   A0  A1 x  a   A2 x  a   A3 x  a    Anx  a 
2 3 n

36
University of Gondar Lecture Notes
 Mathematical Economics

p x   A1  2 A2 x  a   3 A3 x  a    nAn x  a 
2 n 1

p x   2 A2 6 A3 x  a    n  1n An x  a 

n 2


p n x   nn  1n  22.1. An

When x=a
pa   A0  f a 

p a  f a 
p a   A1  1! A1  A1  
1! 1!
p a  f a 
p a   2 A2  2! A2  A2  
2! 2!
p a  f a 
p a   6 A3  3! A3  A3  
3! 3!

p n a  f n  a 
p n a   n n  1n  2   2.1. An  n! An  An  
n n!

f a x  a  f a x  a  f n a x  a 

2 n
 f x   f a    
1! 2! n!

e.g Find the third order approximation of f x   1  x about x=0?

Solution: f 0  1

f x   1 1  x  2  f 0  1
1

2 2

f x    1 1  x   f 0   1
3
2
4 4

f x   3 1  x  2  f 0  3
5

8 8

f 0x  0 f 0x  0 f 0x  0

2 3
Therefore, f x    
1! 2! 3!

f x   1  1 x  1 x 2  1 x 3
2 8 16

37
University of Gondar Lecture Notes
 Mathematical Economics

Self test Exercise 2.2:

1. Find the quadratic approximation for f x   5x  3 , for x close to 0.
2

2. Find quadratic approximation of F(K)=AK  about K = 1 ?

3. Prove that 1  x   1  mx for x close to zero, and use this approximation

m

to the following numbers?


a) 3 1.1  1  1
10
1
3
c) (1.02) 25

b) 5

33  2 1  1
32
 1
5
d) 37  36  1

2.6 Expansion of Functions using Maclaurin and Taylor

series
a)Maclaurin series of a polynomial function (Expansion around the point
x=0)
- Consider expanding an nth degree polynomial function:
f x   a0  a1 x  a2 x 2  a3 x 3    an x n

- Successive differentiation of f x  yields:

f x  a1  2a2  3a3 x 2    nan x n1

f x   2a 2  6a 3 x    nn  1a n x n 2


f n 
x   nn  1n  221a n
Evaluate at x =0
f 0 
f 0   a 0  a 0 
0!
f 0 
f 0   a1  a1 
1!
f 0 
f 0   2a 2  a 2 
2!

f n 
o   nn  1n  2 21a n  n! a n  a n 
f n 
n 
n!
Now, substituting these values into the original function we have.

38
University of Gondar Lecture Notes
 Mathematical Economics

f 0 f 0x f 0x 2 f n  0x n

f x      
0! 1! 2! n!

Example -1: Find the Maclaurin series for the function: f x   8  36x  12x 2  x 3

Solution: f 0  8
f 0  36
f x   36  24x  3x 2
f 0  24
f x   24  6 x f 0  6
f x   6
Thus, the Maclaurin series is:
f 0x f 0x 2 f 0x 3
f x   f 0   
1! 2! 3!
 f x   8  36x  12x 2  x 3  This verifies that the Maclaurin series correctly
represents the given function.
b) Taylor series of a polynomial function (Expansion around any point x=xo)
For the purpose of expansion around a specific point x0, we may first interpret any given
value of x as a deviation from x0.
Let X  X 0   , where  represents this deviation so that   x  x0 .

Consider the function: f x   2  4 x  3x 2

f x   2  4x0     3x0   
2

f x   4  6x0   

f x   6
- Since xo in the present context is fixed, the only variable in the expression
x=xo+  is  . Therefore, we can rewrite f(x) as a function of  as:
g    2  4 X 0     3 X 0     f x 
2

g    4  6 X 0     f x 

39
University of Gondar Lecture Notes
 Mathematical Economics

g    6  f x

- Expanding g(  ) around zero (  =0) yields the following Maclaurin series

g 0 gu0
2

g    g 0  
1! 2!
 The fact that   0 implies X=X0, hence
g 0  2  4 x0  3x02  f  x0 

g 0  4  6 x0  f  x 0 

g 0  6  f x0 

substituting these in to the function g   ,we have

f x0 x  x0  f x0 x  x0 

2

f x   g    f x0   
1! 2!

Example:- Find a Taylor series expansion for the function f x   5  3x  2 x 2  4 x 3 ,

around x0=3.

Expansion of an arbitrary function: From Taylor’s Theorem so far, we have seen how
nth degree polynomial functions can be expressed in another equivalent nth degree
polynomial form. But it is also possible to express any arbitrary function f(X).(one that is
not necessarily polynomial) in a polynomial form.
- According to Taylor’s Theorem, given an arbitrary function  x , which is
continuous with a finite power n, its expansion around the point X0
yields
  X 0    X 0  X  X 0    X 0  X  X 0 2  n   X 0  X  X 0 2
 x         Rn
0! 1! 2! n!
or   X   Pn  Rn , where Pn= the polynomial approach X2
Rn= the remainder.
Examples
1) Expand the non – polynomial function:

 x  
1
, Around the point x 0  1, with n = 4?
1 x

40
University of Gondar Lecture Notes
 Mathematical Economics

We need the first four derivatives;

 x  11  x2 So that  1   1 4

 x   21  x 3  1  1 4

 x  61  x4  1   3 8

 4 x  241  x5  4  1  3 4

And  1  1
2

 x   12  1 4 x  1  18 x  12  116 x  13  13 2x  14  R4

 x   3132 
13
x  1 x 2  3 x 3  1 x 4  R4
16 2 16 32

2) Expand the quadratic function  x   5  2 x  x 2 , around X0 =1? , with n = 1?

 x   8  4x  1  R1  1  5  2  1  8

 1  2  21  4
Lagrange form of the Remainder
- According to Lagrange form of the remainder, we can express Rnas:
 n 1 c 
Rn  x  x0 n1
n  1!
Where C is the value at which the remainder is evaluated and lies between X &
X0.
(For the detail, Read A. Chiang)

2.7 Intermediate Value Theorem (mean)

Let f(x) is a function that is continuous for all x in the closed interval a,b, and assume

that f (a)  f b

41
University of Gondar Lecture Notes
 Mathematical Economics

f(b)
f(x)

f(c)

f(a)
x
a c b

The theorem implies that, if f(x) is continuous, the graph of f(x) must intersect the line
y= f(c) at least at one point. i.e., point (c,f(c)), where a  c  b .
An important implication of the theorem is the following. Let f(x) be a function
continuous in [a,b] and assume that f(a) &f(b) have different signs. Then there is at least
one point c a, b such that f(c) =0.

Example : prove the equation X 6  3x 2  2 x  1  0, has at least one solution between 0

and 1?
Solution: - Evaluate the function at the two extreme values
 f (0)  1
f (1)  1
So, According to the Intermediate Value Theorem, there is at least one solution
c 0,1 such that f(c) =0
We can further narrow down the interval in which the solution is found. For example,
if we take the mid-point of the interval (0,1)= ½

 2  0, Therefore, there is at least one solution in 12 ,1

f 1

 Take, the mid-point of 1 ,1 i.e, x  3 . again f 3   0, implying that there is

2 4 4
a solution between ¾ &1.
Newton’s Method
The Intermediate Value Theorem shows that a given equation f(x) = 0 has a solution
in a given interval. However, it doesn’t provide additional information about the
location of the solution, Newton’s method leads to a better approximate solution.

42
University of Gondar Lecture Notes
 Mathematical Economics

f(x)

f(x0) (x0,f(x0))

f(x1) (x1,f(x1))
a

x0 x1 x2

If f(x)=0 at point a, estimate a by starting an initial estimate x0 of a.

By constructing a tangent line at (x0,f(x0)), we obtain x1. Hence, x1 is a better
approximation of ‘a’ then x0. Repeat the procedure by constructing another tangent
line at the point x1 , f x1  and then find x2.
We can easily find the formula (equation) for Xi’s derived by Newton’s Method
Example 2: What is the equation of the line is given by:
y  y0
m
x  x0

 y  y 0  mx  x0 

 y  f x0   f x0 x  x0 

At x=x1 , y=0
 0  f x0   f x0 x1  x0 

f ( x0 )
 x1  x0 
f x0 

f x1 
x 2  x1 
f x1 

f x2 
x3  x 2  , etc In general , the
f  x12 
Point generated by Newton’s method are given by:

f xn 
xn 1  xn 
f xn 

43
University of Gondar Lecture Notes
 Mathematical Economics

Example 2: - find an approximate value of ‘a’ for the equation:

f x x 6  3x 2  2 x  1  0, in the interval 0,1, using Newton’s mehtode once

f ( x0 )
Solution: Here n = 0 and xo =1 or o (use either extreme) x1  x0  , where
f x0 

f ( x)  6 x 5  6 x  2
f ( x)  10
f ( x)  f (1)  1

 x1  1  1  0.9
10
*(check for f(x1)=f(0.9)=0.1614 which is close to zero.)
Apply Newton’s Method twice?
f x1 
Solution: x 2  x1 
f x1 

f 0.9 
 x 2  0.9   0.028
f 0.9 

2.8 Multivariate Calculus

Partial Derivatives
Given a function y  f ( x1 , x 2  X n ), if the variable x1 undergoes a change

x1 while all other independent variables remain fixed,

The difference quitient in this case can be expressed as:
y f x1  x1 , x 2  x n   f x1 , x 2  x n 

x1 x1

y y
Thus,  f1  im First order partial derivative
x1 x 0 x
1

Second order direct partial

Z  f x, y 
Z
f xx   f x   f xx x, y 
x
Z
f yy   f y   f yy x, y 
y

44
University of Gondar Lecture Notes
 Mathematical Economics

Second order mixed partials (cross – partials)

 Z 
 
f x x   2 Z
    f xy
y 2y  2 2x

 Z 
 
f y  y   2 Z
   f xy
x 2x x 2 y
Young’s Theorem
It states that the mixed (or cross –partials ) for a given function will always be equal if
both cross partials exist. i.e.
Given Y=f(x1,x2)
f12=f21

Ex-1 consider ux1 , x2   x1nx2

Solution :-
u1  nx2

u 2  x1
x2

u12  1
x2

u 21  1
x2

 u12  u 21

Ex-2: for y  f x1 , x2 x1e x1  x2 , show that f12  f 21 ?

2

Solution: f1  x1  1e x1x2 f12  x1  12 x 2 e x1  x2

2 2

f1  x1  12 x2 e x1x2

2
x1  x 22
f 2  2 x1 x 2 e

Generalization of Young’s Theorem:

Given y  x1, x2  xn 

f123 n  f 213 n  f 312 n   f n 312

45
University of Gondar Lecture Notes
 Mathematical Economics

2.4.3. The Multivariate Chain Rule

Many economic models involve composite functions. Differentiation of composite
functions requires the application of chain rule.
-suppose Z  F x, y  and x  f (t ), y  f (t )
dz z x z y
 .  .
dt x z y t
Example:
1) y  nx1  x2 , where x1  2t , and x2  t 2

dy y x1 y x 2
 .  .
dt x1 t x 2 t

2 2t
 
x1  x2 x1  x2
2t  1

2t  t 2

Or by direct substitution: y  n2t  t 2 

dy 2  2t
 
dt 2t  t 2

Suppose z  F ( x, y )  x 2 e y
dz
x  nt, y  t 2 , find ?
dt
The partial derivative case.
Given z  F ( x, y ), where x  f (s, t ), y  g s, t 
z z x z y
 .  .
s x s y t
z z x z y
Similarly:  .  .
t x t y t

Example:
Suppose z  F x, y   x 2  2 y 2

46
University of Gondar Lecture Notes
 Mathematical Economics

Where x  t  s 2
y  ts
z z x z y
 .  .
t x t y t
= 2 x.1  4 y.s

 2t  s 2   4ts .s

 2t  2s 2  4ts 2

2.9 Homogenous functions & Euler’s Theorem

Definition; A function y  f x1 , x 2 , x n  is said to be homogenous of degree r if and

only if f tx1tx2 txn   t r f x1 , x2  xn   t r y, where r measures the degree of

homogeneity. In economic application the constant r is usually taken to be positive, as
most economic variables do not admit negative values.
Examples: Determine whether the following functions are homogenous. If so, of what
degree?

1) f  x, y, z  
x 2z

y 3x

Solution: f tx, ty, tz  

tx 2tz

ty 3tx

t  x 2z  0  x 2z 
    t   
t  y 3x   y 3x 
 t 0 f  x, y , z 

 Here f is homogenous of degree 0(Ho).

2) y  f L, K   L K 1

Solution: f tL, tK   tL tK 

 1

 t  L .t 1 K 1
 tL K 1  t 1 f L, K   H 1

3) f x, y   3x 2 y  y 3

47
University of Gondar Lecture Notes
 Mathematical Economics

 
Solution: f tx, ty  3 tx 2 ty   ty 
3

 3t 2 x 2 ty   t 3 y 3

 
t 3 2 xy 2 y  y 3  t 3 f x, y   H 3

4) f ( x, y )  x 3  xy  y 3
Euler’s Theorems
If f x1 , x 2 , x 2  x n  is homogenous of degree r, then:

r f f
.x1  .x2   .xn  rf x1 , x2  xn 
x1 x2 xn
Proof: by the definition of homogeneity:
f tx1 , tx2 txn   t r f x1 , x2 , xn 
-Differentiating both sides of the identity with respect to t:
f tx1  f txn 
.  .  rt r 1 f x1 , x2  x n 
tx1  t txn  2t

f f
 .x1   .xn  rt r 1 f x1 , x2  xn 
tx1  txn 

Putting t = 1, results the Euler’s Theorem:

f f f
.x1  .x 2   .xn  rf x1 , x2  xn 
x1 x0 xn

Example Find the degree of homogeneity of : f x, y   x 4  x 2 y 2 using Euler’s

Theorem?
- According to Euler’s theorem:
f x X  f yY  rf x, y 

f x  4x 3  2x 2 y 2 , and f y  2yx 2

  
 f x X  f y Y  4 x 3  2 xy 2 x  2 yx 2 y 

48
University of Gondar Lecture Notes
 Mathematical Economics

 4x 4  2x 2 y 2  2 y 2 x 2

 4 x 4  4 x 2 y 2  4 f  x, y  , where r = 4

Extension of Euler’s Theorem:If f x1 , x 2  x n  is homogenous of degree r, then the

first partials f1 , f 2  f n are homogenous of degree r-1.

Proof: By the definition of homogeneity:
f tx1 , tx2 txn   t r f x1 , x2  xn 
- Differentiating both sides of the identity with respect to Xi:
f txi  r f
. t
xi  xi xi

f f
 .t  t r
txi  xi
- Dividing both sides by t:
f f
 t r 1
txi  xi

Linear Homogeneity
An important special case of homogeneity is that of homogeneity of degree one,
which is often referred to as liner homogeneity. Linearly homogenous fucntins are
widely applied in the theory of production and imply the economic assumptionof
constant returns to scale. Because linear homogeneity implies raising all inputs t-
fold, it will always raise output exactly t- fold.
Properties of Linearly homogenous production functions
Given the Linearly homogenous production function Q  f L, K   L K 1 :
1) The average products of each factor can be expressed as functions of the two
input ratios.
1
Q L K 1 K 1  K 
APL    1   
L L L L

49
University of Gondar Lecture Notes
 Mathematical Economics

Q L K 1 K   K 
APL       
K k L L
2) Similarly, the marginal products can be expressed as functions of the ratio of the
two inputs.
1
Q K 1 K
i.e. MPL   L 1 K 1  1    
L L L

Q
MPk  1   L k 
k

k
 1    
 L

Therefore, the marginal products also depend only on ratios of the two inputs.
Conclusion: If labour & capital are changed by the same proportion, the marginal
and average products of the inputs will not be affected.
Example:consider a firm with a linearly homogenous production function:
Q  f L, K 
By Euler’s theorem
f L L  f K K  Q  Economically, this property means that under conditions of
constant returns to scale, if each input is paid the amount of its marginal product,
the total product will be exactly exhausted by the distributive shares fro all the
input factors, or equivalently, the pure economic profit will be zero.
From
Q  f L L  f K K  MPL L  MPK K , Dividing both sides by L, we have
Q K
 MPL  MPK .
L L
K
APL  MPL  MPK .
L

 MPK K  APL  MPL

L

Assuming K L is positive:

50
University of Gondar Lecture Notes
 Mathematical Economics

i) when APL is rising (MPL>APL)  MPK is negative (excess capital)

ii) When APL is falling (MPL<APL)  MPK is positive (capital is utilized)
Exercise 2.3
1. Consider two goods X1 and X2 with prices P1&P2. Suppose that a consumer with
money income M has the following demand function for X1:
MP2
X1 
P12
a. Show the effect of balanced or neutral inflation on the demand of those goods?
b. Show whether Euler’s Theorem holds for this demand function?
2. Find the Maclaurin series for the function: f x   6  8 x  7 x 2  3x 3  2 x 4

51
University of Gondar Lecture Notes
 Mathematical Economics

Chapter -3
Unconstrained Optimization
Aims and objectives:
After successfully completing this chapter, students are expected to:
 Know how to solve Functions of one variable where there is no
constraint
 solve Functions of several variables

Introduction
When finding the relative extrema of objective functions of more than one choice
variable which are independent of one another), the decision made regarding one
variable does not affect the choice of the other variable. Such functions which involve
no constraints are called unconstrained functions & the optimization process is called
unconstrained or free optimization.
From the point of view of economics, extreme (equilibrium) points may be of two
types. These are goal equilibrium and non-goal equilibrium. The former is
equilibrium point which the economic agent is deliberately striving to achieve, and
entails maximization of any form of benefit and minimization of costs (of any form).
The general term for this type of economic problem is optimization. In this chapter,
we will focus on optimization problems using the classical method of optimization
(the concept of derivatives)

3.1. Functions of one variable

Minimum and maximum values of a function occur either at stationary points or at
end points. So, to find the extreme values of a continuous function, the 1st
requirement is to find stationary points. This is called 1st order condition and tested
using 1st derivatives. i.e.

52
University of Gondar Lecture Notes
 Mathematical Economics

However; the 1st order condition is not sufficient to say a function achieves maximum
or minimum, because it tells us only that the function has stationary point (zero
slopes). The sufficient condition is the rate of change of the slope (the 1st derivative).
Given a two variable function y = f(x), the extreme values can be fixed using the
following general rule.
F.O.C: The derivative version The differential version

 f  x   0
dy
dy  f x dx  0, dx  0
dx
d2y
 f CV   0  min
S.O.C : dx 2 d 2 y  f CV dx 2  min
 0  max
> 0 Max
=0 test inconclusive
Example

3.2. Functions of several independent Variables

Given z  f x, y  :
Using the differential version of optimization, we have:
1. F.O.C:
dz  f x d x  f y d y  0,. for.dx  0.and.dy  0
 dz  0, iff fx and fy are zero
Therefore, the F.O.C. will be, fx=fy =0
2. S.O.C; once the 1st order condition is
satisfied, the next step is to check the 2nd order (sufficient) condition. Given
dz  f x d x  f y d y ,

 dz dx  dz dy

d 2 z  d dz   
x y
  f x d x  f y d y d x   f x d x  f y d y d y
 
x y

53
University of Gondar Lecture Notes
 Mathematical Economics

  f xx d x  f yx d y d x   f xydx  f yydy d y
.
f xy  f yx
But by young’s theorem . Substituting this value, we get the following
equation.
d 2 z  f xx dx 2  2 f xy dxdy  f yy dy 2

For any dx and dy, both different from zero, the following rule is used to identify
the maximum and minimum points:
 d 2 z  0  Maximum
 d 2 z  0  Minnimum
But for any values of dx and dy, not both zero:
 d 2 z  0, iff . f xx  0, f yy  0,.and. f xx f yy   f xy 
2

 d 2 z  0, iff . f xx  0. f yy  0,.and. f xx f yy   f xy 
2

So, the conditions for optimum values of a two variable function can be summarized
as follows.
Conditions Maximum Minimum
F.O.C fx  fy  0 fx  fy  0
S.O.C f xx  0, f yy  0 f xx  0, f yy  0

f xx f yy   f xy  f xx f yy   f xy 
2 2

Violation (failure) of the S.O.C

If the second order conditions for maximum and minimum are not fulfilled, the
points will be either of the following.
f xx f yy   f xy 
2
i) If and f xx and f yy have different signs, then the point is
saddle point
f f   f xy 
2
ii) If xx yy and f xx and f yy have the same signs, then the point is
inflection point
iii) If f xx . f yy   f xy  , Test inconclusive
2

Note: - always, we test a point for second order condition only if it passes the
first order condition (first order partial derivative test)

54
University of Gondar Lecture Notes
 Mathematical Economics

Self test Exercise

For each of the following functions, find the critical values, and determine
whether the function is at a relative max, min, an inflection or saddle point?
1) z  f x, y   3x 2  2 y 2  sy  4 x  7 y  12

2) z  f x, y   x  2 y  e x  e 2 y

3) z  e 2 x  2 x  2 y 2  3
Determinant test for the S.O.C
The expression for d 2 z in our derivation of second order total differential
exemplifies what is known as quadratic forms.
Definition a form is a polynomial expression in which each component term has
a uniform degree. i.e. the sum of exponents in each term is uniform.
Example:- f x, y, z   ax  by  z  a.linear. form

f x, y   ax 2  2bxy  Cy 2  a.quadratic. form, etc.

Second order Total Differential as a Quadratic form
Consider d 2 z  f xx dx 2  2 f xy dxdy  f yy dy 2

Let d 2 z  q f xx  a

dx  u f yy  b

dy  V f xy  f yx  h
If we further consider dx & dy as variable and the partial derivatives as
coefficients, then: q  au 2  2huv  bv 2
Algebraic Method and Second Order Condition
Having represented the second order partial differentials as quadratic forms, let us see
how to determine maximum and minimum points using a relatively easy method.
In the case of optimization with one variable, it is easy to test both 1 st order and 2nd order
conditions. But in the case of multi-variable functions, it may be very complex to test the
second order conditions. The complexity increases as the number of variables increase
since we will have many direct and cross partials. In such cases, we resort to the concept
of matrix algebra to simplify our computation.

55
University of Gondar Lecture Notes
 Mathematical Economics

Do you remember sign definiteness of matrices? good. If you forget it, please revise your
Quantitative Method for Economists’ II module. The concept of linear algebra is used in
the following way.
The second- order sufficient condition for extremum stipulates d 2 z to be definitely
positive (for a minimum) and definitely negative (for a maximum)
The major question, then is that: what restrictions must be placed upon a, b and h, while u
and v are allowed to take any values (so long as they are not both zero), in order to ensure
a definite sign for q?
Positive and Negative Definiteness
A quadratic form q is said to be:
- Positive definite, if q  0  S .0. sufficient condition for Min.
- Positive semi-definite , if q  0  S .0 necessary condition for Min
- Negative definite , if q  0  S .0. sufficient condition for Max
- Negative semi-definite, if q≤0. This is a necessary condition for Max
Note:- If q changes signs when the variables assume different values, on the other
hand, q is said to be indefinite (an indication of a saddle point).
Determinant test for sign Definiteness
q  au 2  2huv  bv 2

 au 2  huv  huv  bv 2
 au  hvu  hu  bvv

u 
 au  hv.hu  bv 
v
a h u 
 u v  
h b   v 
a h
Here the determinant of is referred to as Discriminant of the quadratic form q
h b
and is denoted by /D/.
In the particular case of the quadratic form d 2 z  f xx dx 2  2 f xy dxdy  f yy dy 2 , the

discriminant is a determinant with the second-order partial derivatives as its elements.

56
University of Gondar Lecture Notes
 Mathematical Economics

Such a determinant is called a Hessian determinant (or simply a Hessian ) In the two
variable case, the Hessian is
f xx f xy
H 
f yx f yy

From this Hessian determinant (or simply a Hessian).| f xx | f xx , represents the first

principal minor of H and is denoted by H 1 .

The second principal minor of H is

 f xx f yy   f xy 
f xx f xy
H2  H 
2

f yx f yy

The sign definiteness of d 2 z is determined by the sign of the two principle minors. Thus,
i) d 2 z  0 (positive definite) iff:

 f xx f yy   f xy   0
f xx f xy
H 1  f xx  0, and. H 2 
2

f yx f yy

ii) d 2 z  0 (negative definite) iff:

 f xx f yy   f xy   0.
f xx f xy
H 1  f xx  0, and. H 2 
2

f yx f yy

iii) d 2 z is indefinite iff:

f xx f xy
H2   0,
f xy f yy

Example : optimize the following functions

z  f x, y   6 x 2  9 x  3xy  7 y  5 y 2

Solution: F.O.C f x  12x  9  3 y  0

f y  3x  7  10 y  0
Solving simultaneously yields the critical values x =1 & y=1
S.O.C f xx  12 , f yy  10 and f xy  3

Determinant Test
12  3
H1  12  0, and. H 2   111  0
 3 10

57
University of Gondar Lecture Notes
 Mathematical Economics

Thus, d 2 z >0 (positive definite).i.e. the equation is minimized at the critical values.

The cased of three variable quadratic form

Given z= f ( x1 , x 2 , x3 )

f11 f12 f13

Given H  f 21 f 22 f 23
f 31 f 32 f 33

 the sign definiteness of d 2 z depends on the signs of the first , second and third
principal minors, thus;
i) d 2 z >0 (positive define) iff:
H 1  0, H 2  0, and. H  0

ii) d 2 z <0 (negative definite) iff:

H 1  0, H 2  0, and. H 3  0 the three principal minors alternate in sign.

Example:
A firm produces three products Q1,Q2 , and Q3. The profit function is given by:
Q1 , Q2Q3   180Q1  200Q2  150Q3  3Q1Q2  2Q2 Q3  2Q1Q3  4Q12  5Q22  4Q32 ,
then optimize the profit function?
Solution:
F.O.C: 1  0 ,  1  180  3Q2  2Q3  8Q1  0

 2  0 ,  2  200  3Q1  2Q3  10Q2  0

 3  0  3  150  2Q2  2Q1  8Q3  0

,

Critical values: Q1*  14.53, Q2*  13.27, and.Q3*  11.79

8 3  2
S.O.C(Second Order Condition): H =  3  10  2
 2  2 8

H 1   8  8  0

58
University of Gondar Lecture Notes
 Mathematical Economics

8 3 Therefore, the profit function is maximized at the

H2   71  0
 3  10 critical values.
H 3  H  520  0

the case of n- variable Quadratic forms

the algebraic method of testing second order condition can be directly extended for the
case of n-variable quadratic forms as follows.
Given z  f x1 , x 2  x n 

F.O.C. f1  f 2   f n  0

f11 f12  f1n

S.O.C. H  f 21 f 22  f 2n
f n1 f n2  f nn

i) Positive Definite (minimum)

H 1  0, H 2  0, H n  0

ii) Negative definite (Maximum)

H 1  0, H 2  0, H 3  0,   1 H n  0  all odd numbered principal minors are
n

negative, and all even numbered ones are positive.

Exercises
1) A firm produces two products x & y. the cost of producing the products is :
c( x, y )  0.04x 2  0.01xy  0.01y 2  4 x  2 y  500. Suppose that the firm sells all its
output at a per unit price of birr 15 for X, and birr 9 for y. optimize the profit function
of the firm.

2) The production function for product Q is given by: Q  f L, K   6 K 3 L . Let

the price per unit of output be birr 0.5. The cost per unit of capital is birr 0.1 and
the price per unit of labor be birr 1. Optimize the profit function of the firm?
3) Consider a profit maximizing firm that sales its output Q at a constant unit price,
P and purchases two inputs X1 & X2 at constant unit factor prices w1&w2
respectively. The firm faces competitive input & output market and the
production function is given by: Q X 1 , X 2 . then optimize the profit function of

59
University of Gondar Lecture Notes
 Mathematical Economics

the firm and provide economic interpretation of the first & second order
conditions.
4) Consider a monopolist firm that produces two products Q1& Q2. Suppose that the
demands facing the firm are as follows:
P1  55  Q1  Q2
P2  70  Q1  2Q2

The total cost function is given by: cQ1 , Q2   Q12  Q1Q2  Q22 , then optimize the
profit function of the firm and determine the optimal levels of output, price and profit.
5) Suppose a monopolist firm sales a certain product Q in three separate markets.
The demands facing the firm are as follows.
P1  63  4Q1
P2  105  5Q2
P3  75  6Q3
The total cost function is: C Q  20  15Q

Where Q  Q1  Q2  Q3
Then optimize the profit function in of the firm, i.e. determine the prices charged, the
quantities sold as well as the optimal profit.

6) Suppose a monopolist is practicing price discrimination in the sale of product Q

by charging different prices in two separate markets. The demand functions in the
two markets are given by:
P1  100  Q1
P2  80  Q2
Suppose the cost function is: C  6Q
Where Q  Q1  Q2
i) a) How much would be sold in the two markets to maximize profit?
b) What are the prices charged?
c) How much is the maximum profit?
ii) If price discrimination is made illegal:
a) How much would be sold in the two markets?

60
University of Gondar Lecture Notes
 Mathematical Economics

b) What prices would be charged?

c) How much profit would be lost?

61
University of Gondar Lecture Notes
 Mathematical Economics

Chapter 4
Constrained Optimization
Aims and objectives:
Up on successful completion of this chapter, students will be able to solve
. One Variable constrained optimization
. Two variable problems with equality constraint
.Inequality Constraints & The theorem of Kuhn and Tucker

Introduction
In economic optimization problems, the variables involved are often required to satisfy
certain constraints. In the case of unconstrained optimization problems, no restrictions
have been made regarding the value of the choice variables. But in reality, optimization
of a certain economic function should be in line with certain resource requirement or
availability. This emanates from the problem of scarcity of resources. For example;
maximization of production should be subject to the availability of inputs. Minimization
of costs should also satisfy a certain level of output. The other constraint in economics is
the non negativity restriction. Although sometimes negative values may be admissible,
most functions in economics are meaningful only in the first quadrant. So, these
constraints should be considered in the optimization problems.
Constrained optimization deals with optimization of the objective function (the function
to be optimized) subject to constraints (restrictions). In the case of linear objective and
constraint functions, the problems are solved using linear programming model. But when
we face a non linear function, we use the concept of derivatives for optimization. This
chapter focuses on optimization of non linear constrained functions.

4.1. One Variable constrained optimization

i) With equality constraint:- optimization of one variable function subject to an equality
constraint takes the form,
Max (min): y  f x , subject.to : x  x

62
University of Gondar Lecture Notes
 Mathematical Economics

The solution for this type of problem is  y *  f x 

iii) Non- negativity constraint:
a) Max(min): y  f x , subjecto : x  0 .
Consider the following graph.

a
b x 0

y *  f 0

In the above graph, y attains its unconstrained maximum at point a(x<0). But when a non
negativity restriction is imposed on the choice variable, the maximum point will be at
x=0.
x  0, At.x  0, f x  0
b)
Y
y* a
max y *  f x 

x Here both the constrained&

x unconstrained max is at
Y point ‘a’  f x   0 .
a

-Both constrained &

unconstrained max coincide,
and f x   0

63
University of Gondar Lecture Notes
 Mathematical Economics

Minimum:- minimization of functions with non-negativity constraint can be viewed

graphically as follows

y y y

b
min
a
In general, the optimum values of a function with non negativity constraint can be
x <0 as
summarized x =0
follows. x x x
if . f x   0, x  0 x 0
i) Max: f x   0
y *  f 0 if . f x  y *0,x f0x  0 *
y f (0)
f x   0 if . f x  f 0,xx 
00 f x   0
ii) Min: f x   0
if . f x   0, x  0
Example:1
Max y  3x 2  7 x  2, subjectto, x  0
In unconstrained operation:
F.O.C. f x   6 x  7  0

x  7
6

But imposing the non – negative constraint we have:

x *  0, and f 0  2
f 0  7  0

4.2. Two variable problems with equality constraint

In the case of two choice variables, optimization problem with equality constraint
takes the form Max / Min : y  f x1 , x 2 , Subjectto : g x1 , x 2   c . This type of
x1 , x2

optimization problem is commonly used in economics. Because, for the purpose of

simplification, two variable cases are assumed in finding optimum values. For
example in maximization of utility using indifference approach, the consumer is
assumed to consume two bundles of goods.

64
University of Gondar Lecture Notes
 Mathematical Economics

Example: Max ux1 , x2 , subjectto p1 x1  p2 x2  M

In this section, we will see two methods of solving two variable optimization
problems with equality constraints.
i) Direct substitution Method:
Direct substitution method is used for a two variable optimization problem
with only one constraint. It is relatively simple method. In this method, one
variable is eliminated using substitution before calculating the 1st order
condition. Consider the consumer problem in the above example.
P2 x2  M  P1 x1
m p1
x2   x1
P2 p2
Now x2 is expressed as a function of x1. Substituting this value, we can eliminate x2 from
the equation.
 Max, u  ux1 , x2 x1 

du u u x 2
  . 0
dx1 x1 x 2 x1

 Mu1  Mu 2   p1   0
 p2 
p
 Mu1  Mu 2 . 1
p2
Mu1 p
  1
Mu 2 p2
Example u  x1 x2 , subjectto , x1  4 x2  120
4 x2  120  x1

x2  30  1 x1
4

 
 u  x1 30  1 x1  30x1  1 x12
4 4
du
F.O.C.( First Order Condition)  30  1 x1 =0
dx1 2

 x1  60, and, x2  15

iii) Lagrange Multiplier Method

65
University of Gondar Lecture Notes
 Mathematical Economics

When the constraint is a complicated function or when there are several constraints,
we resort to the method of Lagrange.
Max : f x1 , x2  Subject to g x1 , x2   c
L  f x1 , x2    c  g x1 , x2  L denotes the Lagrange.
An interpretation of the Lagrange Multiplier
The Lagrange multiplier,  , measures the effect on the objective function of a one unit
change in the constant of the constraint function (stock of a resource) . To show this;
given the Lagrangian function;
L  f x1 , x2    c  g x1 , x2 
F.O.C.
L1  f 1 x1 x 2   g1 x1 , x 2   0
L2  f 2 x1 x 2   g 2 x1 , x 2   0
L  c  g x1 , x 2   0

We can express the optimal choice of variables * , x1* and. y 2* as implicit functions of
the parameter c:
x1*  x1* c 
x 2*  x 2* c 
*  * c 

Now since the optimal value of L depends on * , x1* .and. x 2* we may consider L*to be a
function of c alone. That is,

 
L*  f x1* (c), x  (c   * (c) c  g x1* (c), x2* (c) , 
Differentiating L* with respect to c, we have:

* d * dx2* 
dL* dx1* dx2*

 c  g x1 , x 2   dx1*
*
 f x1  f x2 *
  1  g x 1  g x2 
dc dc dc dc  dc dc 

d*

 f x1  gx1 g x 1
*
 dx1*
dc

 f x2   g x2
*

dx2*
dc
  
 c  g x1* , x 2*
dc
 *

The expressions with a sign of cancellation are all equal to zero.

dL*
 *
dc

66
University of Gondar Lecture Notes
 Mathematical Economics

Note :- If   0, it means that for every one unit increase ( decrease ) in the constant of
the constraining function, the objective function will decrease ( increase by a value
approximately equal to  .

If   0, a unit increase (decrease) in the constant of the constraint will lead to an

increase (decrease ) in the value of the objective function by approximately equal to  .
S.O.C. Conditions for a constrained optimization problem
Max : f x, y  subjectto, g x, y 
L  f x, y   g x, y 
F.O.C.
L  g  x, y   0 x
L x  f x  g x  0 y
L y  f y  g y  0 

S.O.C: L  0, Lx  g x , L y  g y

Lxx  f xx  g xx , Lxy  f xy  g xy

Lyy  f yy  g yy
In matrix form:
 L Lx Ly   0 gx gy 
   
 Lx Lxx Lxy    g x Lxx Lxy 
L Lyx Lyy   g y Lyx Lyy 
 y

This matrix is called Bordered Hessian and its determinant is denoted by H .

L xx L xy
The bordered Hessian H is simply the plain Hessian   bordered by
L yx L yy

the first order derivatives of the constraint with zero on the principal diagonal.

Determinant Criterion for the sign definiteness of d 2 z .

67
University of Gondar Lecture Notes
 Mathematical Economics

0 gx gy
 Positive.definite   0  min
d 2 z.is  subject.to.dg  0.iff g x Lxx Lxy
 Negative.definite  0  Max
gy L yx L yy

n- variable case
Given f x1 , x 2  x n 

Subject to g x1 , x 2  x n   c

0 g1 g 2  g n
g L11 L12 L1n
H  1 Its bordered leading principal minors can be defined as:

gn Ln1 K n2 Lnn

0 g1 g2 g3
0 g1 g2
g L11 L12 L13
H 2  g1 L11 L12 , H 3  1 etc.
g2 L21 L22 L23
g2 L21 L22
g3 L31 L32 L33

Conditions Maximization Minimization

F.O.C. L  L1  L2  Ln  0 L  L1  L2  Ln  0

S.O.C. H 2  0, H 3 |  0 , H 2 , H 3  H n  0

Example: Max : zx, y   2 xy subjectto 3x  4 y  90

L  2 xy   90  3x  4 y 
The subsequent equations can be solved by either of the following ways.
a) direct substitution
b) inverse method
c) Cramer’s rule
d) Gauss-Jordan Elimination Method
L  90  3 x  4 y  0 x *  15
F.O.C:- L x  2 y  3  0 y *  11.25
L y  2 x  4  0 *  7.5

68
University of Gondar Lecture Notes
 Mathematical Economics

 0 g1 g2  0 3 4
 
S.O.C:  g1 L xx L xy   H  3 0 2
g L yx L yy  4 2 0
 2

H  H 2  48  0  Negative definite ( Max )

More than one Equality constraint

Max / Min : f x1 , x2 , x3 , subjectto , g 1 x1 , x2 , x3 ,  c1and

g 2 x1 , x2 , x3   c 2

   
L  f x1 , x2 , x3   1 c1  g 1 x1 , x2 x3   2 C 2  g 2 x1 , x2 x3 

F.O.C.: L1  f1  1 g11  2 g12  0

L2  f 2  1 g 12  2 g 22  0

L3  f 3  1 g 31  2 g 32  0

L 1  C 1  g 1 x1 , x 2 , x3   0

L 2  C 2  g 2  x1 , x 2 , x3   0

0 0 g11 g 12 g 31
0 0 g12 g 22 g 32
S.O.C. H  g11 g12 L11 L12 L13
g 12 g 22 L21 L22 L23
g 31 g 32 L31 L32 L33

H 2  is one that contains L22 as the last element of its principal diagonal

H 3  is one that contains L33as the last element of its principal diagonal.

A Jacobian Determinant
A Jacobian determinant permits testing for functional dependency for linear & non –
linear functions. A Jacobian determinant is composed of all the first order partial
derivatives of a system of equations arranged in ordered sequence.

69
University of Gondar Lecture Notes
 Mathematical Economics

Example given y1  f1 x1 , x2 , x3 

y 2  f 2 x1 , x 2 , x3 

y 3  f 3 x1 , x 2 , x3 

y1 y1 y1

x1 x 2 x3
y y 2 y 2
J  2
x1 x 2 x3
y 3 y 3 y 3
x1 x 2 x3

- If J  0, then there is functional dependency among equations and hence no

unique solution exists.

y1 y
E.g y1  4 x1  x 2  y11  4 & y 12  1 y 12 means and y11 means 1
x 2 x1

y2
y2  16x12  8 x1 x2  x22   32x1  8 x2
x1

y 2
 8 x1  2 x 2
x 2

4 1
J   64x1  16x2  0, thus, the system has a solution because the
32x1  8 x2 8 x1 2 x2

value of the Jacobian determinant is non-singular, i.e., different from zero.

As stated in the preceding paragraphs, partial derivatives can provide a means of

testing whether there exists functional (linear or nonlinear) dependence among a set
of n functions in n variables. This is related to the notion of Jacobian determinant.
Exercise
1. Optimize: y  3x12  5 x1  x1 x2  6 x22  4 x2  2 x2 x3  4 x32  2 x3  3x1 x3, using.

a) The Cramer’s rule for the first order condition.

b) The Hessian for the S.O.C.
2. A monopolistic firm produced two related goods. The demand and total cost
functions are:

70
University of Gondar Lecture Notes
 Mathematical Economics

Q1  100  2 p1  2 p2
Q2  75  0.5 p1  p2

TC  Q12  2Q1Q2  Q22

Optimize the firms profit function by:
a) Finding the inverse demand function.
b) Using cramer’s rule for the f.o.c.
c) Using the Hessian for the s.o.c.
3. Maximize a firm’s total cost:
c  45x 2  90xy  90 y 2 , when the firm has to meet a production quota:
2 x  3 y  60, by:
a) finding the critical values
b) using the Bordered Hessian to test the S.O.C.
4. A monopolist firm produces two substitute goods and faces the following
demand functions.
X  80  Px  2 p y

y  110  2 px  6 p y

Its total cost function is : TC  0.5 x 2  xy  y 2 what is the profit maximizing level
of output if the firm has to meet a production quota of 3 x  4 y  350?

5. Maximize utility: u x, y   x 0.25 y 0.4 , subjectto, 2 x  8 y  104 (use any method).

4.3. Inequality Constraints & The theorem of Kuhn and

Tucker
Inequality constraints are those requiring certain variables to be non-negative. These
often have to be imposed in order that the solution should make economic sense. In
addition, bounds on the availability of resources are often expressed as inequalities
rather than equalities. Consider the following examples.
Ex-1: Max :   5x1  3x2 ,  stands for profit.

71
University of Gondar Lecture Notes
 Mathematical Economics

6 x1  2 x 2  36
5 x1  5 x 2  40
subject to  Here we add slack variables to make the constraint binding
2 x1  4 x 2  28
x1 , x 2  0 

(active)
2 x1  x 2  14
x1  x 2  12 
Example-2: Min : C  2 x1  4 x 2 : subject to  Here we subtract surplus
x1  3x 2  18
x1 , x 2  0 

values to make the constant binding. C denotes cost.

Note :- If at optimality the slack/surplus of a constraint is zero, then the constraint is

said to be active or binding. Otherwise, it is inactive.

Effects of non – negativity Restrictions

i. single variable case
Max :   f x1 , subjectto , x1  0
In view of the restriction x1  0, three possible situations may arise.

  f x1    f x1 
A

B C
D
E
П=f(x1)

X1 X1 X1

f x1   0, and.x1  0 po int . A These three conditions can

be consolidated into a single
f x1   0, and.x1  0 po int .B statement:
f x1   0, and.x1  0 po int .c & D f x1   0, x1  0.and.x1 f x1  (
for max) and

f x1   0, x1  0, and.x1 f x1   0 for. min imun

72
University of Gondar Lecture Notes
 Mathematical Economics

- the equation x1 f x1   0 is referred to as the complimentary slackness between

x1 & f x1 .
ii. Two variable, one constraint case
max   f x1 , x2 , subjectto, g x1 , x2   r
x1 , x2  0
And slack variable to change the constraint into strict equality, thus .
Max   f x1 , x2 , subjectto, g x1 , x2   s  r
x1 , x2 , s  0
Step1: Form the Lagrangian function
L  f x1 , x2   r  g x1 , x2   s
Step2:Find the first order partials with respect to each of the choice variables and set
them equal to zero.
L
L x1  0, x1  0.and.x1  0  1
x1
L
L x2  0, x 2 , and.x1  0  2 
x 2
L
Ls  0, s  0, and.s  0  3
s
L  r  g  x1 , x 2   s  0  4 
Now, we can consolidate equation (3) and (4) and in the process, eliminate the dummy
L
variables. i.e.    , thus equation (3) tells us that    0, s  0, and  s  0, or
s
equivalently
  0, s  0, and.s  05
 From equation(4) we have
s  r  g x1 , x2 
 substituting (4) in (5) we have
  0, r  g x1 , x2   0, and.r  g x1 x2   0
- thus, the Kuhn-Tucker( K-T) conditions for maximization problem with
inequality constraint are:

73
University of Gondar Lecture Notes
 Mathematical Economics

L1 L
 f1  g1  0, x1  0, and.x1 0
x1 x1

L2 L
 f 2  g 2  0, x 2  0, and.x 2 0
x 2 x 2

L
 r  g  x1 , x 2   0,   0, and. r  g  x1 , x 2   0

- Kuhn-Tucker(K-T) conditions for minimization problem with inequality
constraint:
Lx1  f1  g1  0, x1  0, and.x1 Lx1  0

Lx2  f 2  g 2  0, x2  0, and.x2 Lx 2  0

L  r  g x1 , x 2   0,   0, and.L  0

[Students’ activity:- For n- variable, m-constraint case, please read A.C .

Chiang]

Example- 1
max y  10x1  x12  180x2  x22 , subject to, x1  x 2  80
x1 , x2  0
Step -1:- Form the Lagrangian function
L  10x1  x12  180x 2  x 22   80  x1  x 2 
Step -2: solve for the first order condition (F.O.C.) and set the partials with respect to
each of the choice variables equal to zero.
L x1  10  2 x 2    0  1
L x2  180  2 x 2    0  2 
L  80  x1  x 2  0  3

From (1) & (2) we have, x2  x1  854

 Substituting (4) in (3), we have x1   5 (not feasible) because the x’s should be
2
non-negative.
 Hence setting x1*  0, we have x2*  80

And *  20

74
University of Gondar Lecture Notes
 Mathematical Economics

Step -3: Check whether the constraints are satisfied

i) non – negativity constraint x1*  0

x 2*  80

*  20
ii) inequality constraint : x1  x2  80
0  80  0
L
Step -4: check the complementary slackness condition  x,  0  since x1is zero,
x1
L
x,  0.
x1
L L
 x2 ,  0, x2*  80  0
x1 x2

180  2 x2    0
180  280  20  0
00
L L
-  0, *  20  0
L 
 80  x1  x2  0
80  0  80  0
00
* Since all the K- T conditions are satisfied, the function is maximized at
x1*  0 & x2*  80.
Example -2
Minimize C  x1  4  x2  4
2 2

2 x1  3x 2  6
Subject to:  3x1  2 x 2  12
x1 , x 2  0
Solution:
Step1: Formulate the Lagrange

75
University of Gondar Lecture Notes
 Mathematical Economics

L  x1  4  x2  4  1 6  2x1  3x2   2  12  3x1  2 x2 

2 2

F.O.C.(First Order Condition)

L1  2x1  4  21  32  0 1
L2  2x 2  4  31  22  0 2
L1  6  2 x1  3x 2  0 3
L2  12  3x1  2 x 2  0 4

To solve such multivariable non –linear programming problem, let us use iteration or
trial & error ), method
Iteration-1 1  0, and.2  0
L L
Thus we have =0 and =0 i.e. 2x1 + 3x 2 =6 and 3x1  2 x2  12
1 2
24 6
Solving simultaneously yields x1  and x 2  (this is not feasible).
5 5
From equation (1) & (2), we have:
2 x1  21  32  8, and 5
2 x2  31  22  86
L
And 2  0   3x1  2 x2  127
2
 solving equations (5) , (6),and (7) assuming 1  0, yields

x1*  28
13
 2  3613
*

1  0
 2  1613

 check whether the constraints are satisfied

i) Non – negative constraint  satisfied
ii) Inequality constraint:

76
University of Gondar Lecture Notes
 Mathematical Economics

 18   36 
2   3   6
 13   13 
 28   36 
 3   2   12
 13   13 
 Check the complementary slackness conditions. This is satisfied.
* Therefore, since the solution values for the four variables satisfy all the K-T conditions,
they are acceptable as the final solutions.

77
University of Gondar Lecture Notes
 Mathematical Economics

Chapter -5
5. Comparative static analysis

After successfully completing this chapter, students should be able to:

 Appreciate the importance of comparative statics in economics
 Understand differentiation and its application to comparative statics
 Analyze Comparative Statics of General function model
 Identify the limitations of comparative static analysis

Introduction
Dear Distant Learner in comparative statics, we simply compare the initial (or pre-
change) equilibrium state with the final (or post change ) equilibrium state by
disregarding the process of adjustment of the variables over times. Thus comparative
statics is essentially concerned with finding the rate of change of the equilibrium value of
a choice(endogenous) variable with respect to a change in a particular parameter
(exogenous) variable.

Student activity- what happens to the quantity demanded of a good when the price of
related goods or income changes?

In general, the logical simulation of the testing of theories in economics is called the
theory of comparative statics.

5.1. Differentiation and its application to comparative static

Analysis
Let R(x) = total revenue function
C(x)= total cost function
t(x)= per unit tax (beyond the control of the firm )

78
University of Gondar Lecture Notes
 Mathematical Economics

Max : x   Rx   Cx   tx

Find the first order condition. We all know that the necessary condition for profit
maximization stipulates that marginal revenue should be equal to marginal
cost.(MR=MC)
x   Rx   C x   t  01
S.O.C  x   0 
 Rx   Cx   02
Now , the profit maximizing level of output for the firm depends upon the tax rate(an
exogenous variable).
i.e. X  X * t  3
substituting (3) in (1):
   
R  x * t   C  x * t   t  0

 R x * .  C x * .
dx* dx*
1  0
dt dt


dx*
dt
 
R x *   C x *   1

dx* 1
  0
dt R x   C x * 
*

Output X will decline as the tax rate the firm faces increases. Thus, a prediction about
changes in the choice variable, i.e. marginal adjustment of output, when the parameter
facing the decision maker changes is easily derived, this is the goal of comparative
statics.

Example - 2 x   p.x  cx 

F.O.C: x  p  cx 1
S.O.C x   cx   0  cx   02

X  X *  p 3
Substituting (3) into (1).
 
p  c x *  p   0

79
University of Gondar Lecture Notes
 Mathematical Economics


dp
dp
 
 c  x *
dx*
dp
0
dx*

1
dp c x 
0

This means that if the output price to a competitive firm is raised, output level is
increased. i.e. the supply function is upward sloping.
Market model
Consider one commodity market model.
D : Q  a  bp, a, b  0
At equilibrium Qd  Qs
S : Q  c  dpc, d  0
With solutions:
ac
P* 
bd
ad  bc
Q* 
bd
What would happen to the equilibrium P&Q, if one or more of the parameters changes?
p * 1 p * 1
 0  0
a b  d c b  d
p *  a  c  p *  a  c 
 0  0
b b  d 2 b b  d 2

Q Q
a  a S  b S
a

D
P D

P*  P P *  P 

Notional Income model

Consider the following simple national income model:
y  c  I 0  G0

C     Y  T ,   0,0    1

80
University of Gondar Lecture Notes
 Mathematical Economics

T    y, r  0,0    1
Where,   non – income tax revenue
y  income tax revenue
  income tax rate.
-Endogenous variable = Y, C, & T
- Exogenous variable = I0& G0
-Exogenous parameters =  ,  ,  .and .
The model can be solved for y*as:
    I 0  G0
y*  , this is the reduced form of the above national income model.
1    
From this equation income value, we can have six comparative static derivatives as we
have six exogenous variables.
y * 1
1.  0
G0 1    

y * 
2.  0
 1    

y *  y *
3.  0
 1    

5.2. Comparative Statics of General function model

In the comparative static problems considered before, equilibrium values of endogenous
variables of the model could be explicitly expressed in terms of the exogenous variables;
accordingly, the technique of simple partial differentiation was all we need to obtain the
desired comparative static information.

However, when a model contains functions expressed in general form, explicit solutions
are not available. In such cases, a new technique must be employed that makes use of
such concepts as implicit function rule to find the comparative static derivatives directly
from the given general function model.

Example :- Consider a market model:

81
University of Gondar Lecture Notes
 Mathematical Economics

D D
Qd  D p1 y 0 ,  0, 0
p y 0
s
Qs  s  p , 0
p
At equilibrium D( P, y 0 )  S ( P)  D( P, y 0 )  S ( P)  0
Where , p= endogenous
Y0=exogenous.
We know that every equation price is a function of income. i.e. p  p *  y0  . Therefore,
the equilibrium condition can be taken to be an identify in the equilibrium solution.
   
 D p * , y0  s p *  0
 F p , y   0
*
0

 The comparative static analysis of this model will therefore be concerned with how a
change in y0 will affect the equilibrium position of the model.
i.e.i) what is the effect of a change in Y0on p*?
F

dp *
y 0  D y 0
  0
dy0 F p *
D * s

p () p * ()

Thus, increase y  P*or the other way.

iii) what is the effect of a change in Y0 on Q* ?
At equation , Q*=Qd=Qs
We can write ,
 
Q*  S P* , and.P  P*  y0 

dQ* ds dp
  *. 0
dy0 dp dy0
Thus, the comparative static results convey the proposition that an up –ward shift of
the demand curve ( due to a rise in income) will result in a higher equilibrium price
as well as a higher equilibrium quantity.

82
University of Gondar Lecture Notes
 Mathematical Economics

S0
P1   y 
P0  P
D1

Q0* Q1 D
0
Q

5.3. Limitations of comparative static analysis

Comparative static is helpful in finding out how a disequilibrating change in a
parameter will affect the equation state of a model. However, by its very nature,
comparative statics has the following limitations.
1) Ignores the process of adjustment from the old equilibrium to the new one.
2) Neglects the time element (length of time ) involved in the adjustment process
from one to another equilibrium
3) Assumes that a new equilibrium can be defined and attained after a
disequilibrating change in a parameter, i.e. disregards the possibility that the
new equilibrium may not be attained ever because of the inherent instability of
the model all these limitation are addressed by dynamic analysis which will be
dealt in the next chapter.

CHAPTER 6
6. Dynamic equilibrium analysis

Aims and objectives:-

83
University of Gondar Lecture Notes
 Mathematical Economics

After completing this chapter students should be able to:

 Demonstrate how a time lag can affect the pattern of adjustment to equilibrium in
some basic economic models.
 Distinguish between continuous time and discrete time models
 Set up and solve linear first-order difference and differential equations.
 Apply the differential equation to find solutions to Domar’s economic growth
model, Solow growth model and Market equilibrium models.
 Apply the difference equation solution method to the cobweb iterative model
 Identify the stability conditions in the above model

Introduction
In the previous chapters much of the economic analysis used has been comparative
statics. This entails the comparison of different (static) equilibrium situations, with no
regard on how variables adjust to their new equilibrium values. The study of the
mechanism by which variables adjust between equilibrium values is called ‘dynamics’.
The way markets adjust to equilibrium over time shows a great variation. If the task of
adjusting to equilibrium from a given point is left for market, the adjustment process will
be completely different from the way static equilibrium is achieved. In some markets, the
adjustment process is very fast (instantaneous). In other markets, the adjustment process
is slow, and even we may come across functions which follow a path which diverge from
their dynamic equilibrium point. But, in static equilibrium, there is no model that can
explain these dynamic adjustments. This chapter will introduce you to such models that
explain dynamic equilibrium.
Dynamic equations may be viewed in two ways. Based on the dynamic behavior of
variables, time may be considered as continuous or discrete quantity. In the language of
Mathematics, the former is called differential equation, and the latter difference equation.

Dear Distance Learners, we think that you have some idea about simple dynamic
equations in your Quantitative Methods course. For this course we will discuss simple
dynamic equilibrium analysis in the cases of both continuous and discrete time. But it
will be a bit deeper than the former.

84
University of Gondar Lecture Notes
 Mathematical Economics

6.1 Definition of differential equation: -

An equation F(x; y) which assumes a functional relationship between independent
variable x and dependent variable y and includes x, y and derivatives of y with respect to
x is called an ordinary differential equation. It is an implicit functional relation ship
between functions and their derivatives.
The order of a differential equation is determined by the order of the highest derivative in
the equation. The highest power to which the highest order derivative is raised is called
the degree of the differential equation.
d2y
Example 1. y  2 xy  0 Is a second order differential equation of degree one.
dx2
d2y 1
Because the highest order derivative in the equation is (2nd order derivative) and
dx 2
the power to which this derivative is raised is 1.
4 6
d3y  d2y 
Example 2  3   y  2   4 x  0 is a 3rd order differential equation of degree 4
 dx   dx 
Any differential equations course will concern itself with answering one
or more of the following questions.
1. Given a differential equation, will a solution exist?
Not all differential equations will have solutions so it’s useful to
know ahead of time if there is a solution or not. We proceed to solving
a differential equation only if it has a solution. This question is
usually called the existence question
2. If a differential equation does have a solution how many solutions are there?
It is possible for a differential equation to have more than one
solution. We would like to know how many solutions there will be for a
given differential equation.

1
d2y
Is an operator symbol indicating 2nd derivative and should not be confused with the power to which
dx 2
the function is raised.

85
University of Gondar Lecture Notes
 Mathematical Economics

It would be nice if, during the derivation of our differential equation,

we could make sure that our assumptions would give us a differential
equation that upon solving will yield a single unique solution.
This question is usually called the uniqueness question in a differential
equations course.
3. If a differential equation does have a solution can we find it?
This may seem like an odd question to ask and yet the answer is not
always yes. Just because we know that a solution to a differential
equations exists does not mean that we will be able to find it.

6.1.2 First order linear differential equation

The first special case of first order differential equations that we will look at in this
module is the linear first order differential equation. In this case, the highest order
derivative appearing in the equation is first order derivative. More over it is linear for that
the highest power to which the derivative is raised is 1.
A first-order linear differential equation is an equation of the form

 p t  y  Q t 
dy
dt
Where P and Q are continuous functions of t. This first-order linear differential equation
is said to be in standard form.

6.2 Methods of solving differential equation

 f  x  solution
dy
1. Direct integration: - for a differential equation of the form
dx
can be found using our knowledge of integral calculus i.e.

 f x 
dy
dx
dy  f x dx

 dy   f x dx , taking the integral of both sides

y  F x   c

86
University of Gondar Lecture Notes
 Mathematical Economics

2. Separable Differential Equations

A separable differential equation is any differential equation that we can write in the
following form.

N y  M x 
dy
dx
Note that in order for a differential equation to be separable all the y's in the differential
equation must be multiplied by the derivative and all the x's in the differential equation
must be on the other side of the equal sign.
The solution for the above type of equation is found first by rewriting it as
N ( y)dy = M ( x)dx
Then integrating both sides we get  N ( y)dy   M ( x)dx
Example 1
Consider the differential equation

M ' t   2M t  t
2

After a simple rearrangement, we may write this equation as

 dM
 2tdt Then, taking the integral of both sides, we get the following
M2
result
dM 1
 2
  2tdt →  t2  c
M M

Finally, our solution set becomes M t   1 2  c

t

Example 2
Consider the differential equation

M ' t   t
xt 4  1

Separating the variables yields

87
University of Gondar Lecture Notes
 Mathematical Economics

M 4

1 dM  tdt
The value of M2 + 1 is not 0 for any value of M, so all the solutions of the
equation are obtained by integrating both sides of this equation, which leads to
M t 5  M t   t 2  c
5 2

2. Exact Differential Equation

Given a function of two variables F(y, t), its total differential is
dF dF
dF ( y, t )  dy  dt . When the left hand side of the equation is set equal to 0, the
dy dt
dF dF
equation becomes dy  dt  0 . ……………. (a)
dy dt
This is called exact differential equation.
Generally, an equation of the form Mdy  Ndt  0 is said to be exact if and only if there
F F
exists a function F(y, t), such that, M  andN  ………….(b)
y t
M N
Using Young’s theorem, this may also be written as  2
t y
Solution to exact differential equation:-
dF ( y, t )  o  F  y, t   c . But the main task is to find the primitive function F(y, t). to
F
find the primitive function, we should resort to integration. Since M  , F(y, t) must
y
contain the partial integral3 of M with respect to y. this leads us to
F  y, t    Mdy   t  .

2F 2F   F    F  M N
      
t  y  y  t 
2
By Young’s theorem,
yt ty t y
3
Partial integration is the opposite of partial differentiation, and in the integration process, all terms that don’t contain the variable of
integration should be considered as constants.

88
University of Gondar Lecture Notes
 Mathematical Economics

Where  t  represents those additive terms containing only t and lost during partial
differentiation with respect to y. the first term on the right side may be easy to find. But
dN
the difficulty is to find  t  . We use the second partial differential to find  t  .
dt

3. Using Integrating factor

To solve a first-order linear differential equation, we can use an integrating factor u(x)
which converts the left side into the derivative of the product u(x)y. that is, we need a
factor u(x) such that
d u x  y 
u x   u x  px  y 
dy
dx dx
u x  y  u x  px  y  u x  y '  yu ' x 
'

u ' x 
px  
ux 
ln u x    px dx  c
p  x dx
u x   ce 
Because you don’t need the most general integrating factor, let c=1 Multiplying the
p  x dx
 p  x  y  Q  x  by u  x   e 
dy
original equation produces
dx
p  x dx p  x dx p  x dx
y' e   ypx e   Qx e 

d   p  x dx  p  x dx
 ye   Qx e  So the general solution becomes
dx  

p  x dx p  x dx
ye   Q x e  dx  c

89
University of Gondar Lecture Notes
 Mathematical Economics

THEOREM 6.1 Solution of a First-Order Linear Differential Equation

 p  x  y  Q  x  is
dy
An integrating factor for the first-order linear differential equation
dx
p  x dx p  x dx
and the general solution is ye   Q  x e 
p  x dx
u x   e  dx  c 4

The use of integrating factor changes the function in to an exact differential equation.

4. Bernoulli equation
A well-known nonlinear equation that reduces to a linear one with an appropriate
substitution is the Bernoulli equation, named after James Bernoulli (1654-1705)

 Pt  y  Qt  y n .
dy
The equation takes the form of
dt
When the power in the right hand side (n) is zero the equation become linear and can be
solved using the ordinary methods. If n is one the equation becomes separable differential
equation discussed above. Since w has discussed these issues above, our focus here will
be when n  0andn  1 .
The solution for the above equation may be found as follows

 Pt  y1 n  Qt  Multiplying both sides by y  n

dy
yn
dt

 1  n  y  n . By substitution we get,
dz dy
Let z  y1 n then
zt dt

 Pt z  Qt 
1 dz
1  n  dt
 1  n P t z  1  n Qt  . Now it is very easy to find the solution
dz
dt
since the equation is liberalized. Using integrating factor, the solution becomes
1 n  p t dt 1 n  p t dt
ze   1  n Qt e  dt  c

1 n  p t dt 1 n  p t dt

y1 n e    1  n Qt e  dt  c

p  x dx
4
You had better memorize the integrating factor u  x   e  used to transform the left hand side
p  x dx
into ye rather than the process.

90
University of Gondar Lecture Notes
 Mathematical Economics

dy
 ty  te t y  3
2
Example 1:- find the general solution for
dt
The equation is Bernoulli equation with n=-3.
dz dy
Let z  y1 n  y1 3  y 4 and  4 y 3 ………….. (1)
dt dt
Multiplying both sides of the original equation by 4 y 3 produces
dy
 4ty 4  4te t Substituting from (1) above
2
4 y3
dt
dz
 4tz  4te t ……… (2)
2

dt
The last equation is linear in z and the value of P(t) is 4t.

 pt dt   4tdt  2t

2
2
. So the integrating factor is e 2t
2
Multiplying both sides of equation 2 by the integrating factor e 2t produces an exact
differential equation of the form
dz 2t 2 dz 2t 2
e  4e 2t tz  4tet . The left hand side e  4e 2t tz can be written in total
2 2 2

dt dt

differential form as
d
dt
ze2t  
2
5

d
dt
 
ze2t  4tet
2 2

d
 dt ze 
2t 2
dt   dt
4te t2
→ taking the integral of both sides.

ze2t   4tet dt
2 2

ze2t  2et  c
2 2

z  2e t  ce 2t
2 2

Finally we should transform the solution we get in terms of z, back to a solution in terms
of y using z  y 4

5
Note:-
d
dt

ze2t   e 2t
2

2 dz

dt
 4te2t z
2

91
University of Gondar Lecture Notes
 Mathematical Economics

y 4  2et  ce 2t is a general solution.

2 2

6.3 First order linear differential equation with constant

coefficient and constant term
To make the basic idea progressively clear, let us start with the simplest case i.e. when
P(t) and Q(t) are both constants. The equation of this type is called a first-order linear
differential equation with constant coefficient and constant term, and takes the form
dy
 vy  z , where Z and V are constants. Still we may have another
dt
classification based on the value that the constants take.
The case of z=0: - in this case the equation takes the following form
dy
 vy  0 . This equation is called Homogeneous Linear differential
dt
equation. The solution can be found easily as follows.
dy dy
 vy →  vdt
dt y
dy
 y 
  vdt

= ln y   vt  c

= y  e vt  c  ec  e vt . Let e c  A

= y  Ae  vt

The above solution is a general solution for the homogeneous equation because the value
of A is not known. To get definite solution we should return back to the initial condition
(the value of y when t=0)
y 0  Ae  v0  A . Substituting this value, we get the definite solution which is

y t   y 0e  vt

The case of z≠0:- in this the equation is called non homogeneous equation, and it is of
dy
the form  vy  z . This equation has two components i.e., the particular integral
dt

92
University of Gondar Lecture Notes
 Mathematical Economics

and the complementary equation. So, to find the solution of the complete equation, we
should find the solution of each of the components.
The solution of the complementary equation is the same to the solution of the
homogeneous equation we have seen above.
To find the solution of the particular integral, we mostly use iteration. We start from the
simplest form that y can take. The simplest form for our case is when y=k where k is a
constant function. In this case the non homogeneous equation become
vy  z
dy
z Since y is a constant function, 0
yp  dt
v
z
So, y p  , is the solution for the particular integral.
v
The general solution for the complete system will be
yt   y p  yc , where yc is the solution for the complementary equation derived from
the homogeneous equation.

y t   Ae  vt 
z
to get definite solution we see the initial value of y at t=0
v

y 0  Ae  v0 
z z
 A
v v
A  y 0 
z
v
 z
Substituting the value for A we get the definite solution yt    y0  e vt 
z
 v v

dy
The case of v=0 ; in this case, the equation becomes  z and the solution is easily
dt
derived using direct integration as
dy
 dt dt   zdt  y  zt  c is the general solution and the

definite solution may be found by substituting c with a definite value using the initial
condition, y0  c
yt  zt  y 0 is the definite solution.

93
University of Gondar Lecture Notes
 Mathematical Economics

6.4 Practical Application of First Order Linear Difference

Equation in Economics
Introduction:- Many economic models with a temporal dimension involve relationships
between the values of variables at a given point in time and the changes in these values
over time. A model of economic growth, for example, typically contains a relationship
between the change in the capital stock and the value of output. When time is modeled as
a continuous variable, this economic relation ships are modeled using differential
equations.
Dear distance learners, in this section we will try to discuss some typical applications of
differential equation in modeling economic relationships. The focus of this module will
be entirely on the mathematical part of the models. The theoretical parts may be found in
your Microeconomics Macroeconomics and Development Economics Modules. So, if
you have any difficulty regarding the theoretical frame work, you would probably consult
the aforementioned modules.
1. Domar’s Growth Model
Premises: - 1. A change in I(t) has a dual effect i.e., both on the demand and supply side
2. I(t) has a multiplier effect on the aggregate demand and the multiplier is
1 1
or 6Where c is marginal propensity to consume and s marginal
1 c s
propensity to save.
Suppose investment is the only parametric expenditure in the economy.
dy dI dy
 .  chainrule
dt dt dI
dy dI 1 dy 1
 .  b/c  …………………… (1)
dt dt s dI s
The effect of investment on the supply side, according to this model, is through the
increase in capacity capital ratio. Let δ be the capacity of the economy (measured as

potential output) & k(t) be capital stock    I t  then;

t

 0 

6
The multiplier may be derived from the IS-LM model of a closed economy by taking investment as an
exogenous variable of time.

94
University of Gondar Lecture Notes
 Mathematical Economics


  Where  (rho) is the capacity capital ratio.
k
  k
d
 I t 
dk dk
If we assume  to be constant   but
dt dt dt
d
 I t  ………………………… (2)
dt
At equilibrium, the aggregate demand should be exactly equal to the potential output
dy d
i.e., y   . Taking the derivative of both sides,   I t 
dt dt
dy dI 1
Substituting  . from equation (1) we get
dt dt s

 I t    sI t   0 ……………..(3)
1 dI dI
s dt dt
The last equation above is a homogeneous first order linear differential equation with the
constant coefficient being s . So the solution becomes

I t   Ae  st generalsolution
I t   I 0 e  st definitesolution

2. Dynamic price adjustment model

Qd  a  bP
From comparative statistics if
Qs  c  dP
At market clearing price level
Qd  Qs
a  bP  c  dp
ac
P
bd
But the value doesn’t mean inter-temporal equilibrium. If the task of adjustment is
left for market with out any interference, the adjustment entirely depends on the
relative strength of supply and demand. We should note, however; that it is not only p
that adjusts; but Qs and Qd too. Let’s assume that the rate of change of price is a
linear function of the excess demand/supply (Qd-Qs); then

95
University of Gondar Lecture Notes
 Mathematical Economics

  Qd  Qs  Where  is adjustment coefficient?

dp
dt

  a  bp   c  dp
dp
dt
  b  d  p   a  c 
dp
dt
Where a,b,c and d are constants. The above equation is a linear non homogeneous
difference equation where v   b  d andz   a  c . So, the general solution becomes

pt   Ae  vt 
z
v
ac
Pt   Ae  b  d t  generalsolution.
bd
 a  c   b  d t a  c
pt    p0  e  definitesolution
 bd  bd
ac
But  p , substituting this value in the above equation, we get the following.
bd

 
pt   p0  p e b  d t  p Where b and d are slopes of demand and supply curves.

The last equation tells us that the inter-temporal equilibrium depends on the slopes of
supply and demand curves.i.e, b+d>0 to get a dynamically stable path (time path which
converges to equilibrium). Actually, for normal demand and supply curves this
requirement is always fulfilled because, since sign is considered in the equation, both b
and d are positive. If the conditions are fulfilled, the dynamic time path will be
convergent.

If p0  p , the time path of the function asymptotically approaches p from above

If p0  p , the time path of the function asymptotically approaches p from below

If p0  p the time path will be a horizontal line at p

3. Solow growth model

Basic Assumptions
1. Continuous time.
2. Single good produced with a constant technology.
3. No government or international trade.

96
University of Gondar Lecture Notes
 Mathematical Economics

4. All factors of production are fully employed.

*
L
5. Labor force grows at constant rate n 
L
6. Initial values for capital, K0 and labor, L0 given.
The neoclassical cobb-duglas production function is given by

Y t   AK t  Lt  ………………………..(1).
a 1 a

The production function assumes a constant return to scale. More over, the marginal
productivities of inputs are positive. I.e.,
Q
 aAK a 1L1 a  0
K
Q
 1  a AK a L a  0
L
And decreasing (diminishing) i.e,

 2Q
 a  1aAK a2 L1a  0
K 2

 2Q
 a1  a AK a La 1  0
L 2

If we define all the terms of the equation in per worker terms;

1 a
Y AK a L1 a
a
 K   L
y   A   
L L  L   L …………………………..(2)
y  Ak a

The capital accumulation in the economy at a given period depends on the net investment
in the period. So capital accumulation grows at a rate investment net of depreciation. If δ
is the rate at which capital depreciates, K δ will be the total investment for replacement of
depreciated capital. So the net investment is given by,
I n  sY  K
*
K  sY  K ………………………………… (3)
Where s is a constant saving rate (by assumption)

97
University of Gondar Lecture Notes
 Mathematical Economics

Dividing both sides of equation 3 by K gives

*
K sY
 
K K

* Y *
s
K K y
 L     s   ……… (4) (Taking the per worker values of the variables)
K K K k
L
* * * * * *
k K L K K k
But     n    n ………………. (5)
k K L K K k
Substituting these values in equation 4;
*
k y
 n  s   , but y  Ak a
k k

k  sAk a    nk
*

k    nk  sAk a ……………………… (6)

*

The last equation is a differential equation and it is in the form of Bernoulli equation
where n=a. the dynamic equilibrium (steady state) can be found by using solution
methods of Bernoulli equation as follows;

k k  a    nk 1 a  sA Multiplying both sides by k  a

*

   n z  sA Substituting z  k 1 a and  . k  1  a k  a k
1 dz dz dz * *

1  a dt dt dk

 1  a   n z  1  a sA ………………………. (7)

dz
dt
The last equation above is the linearized form of the Bernoulli equation where a, s, δ and
A are constants. So, the solution is

z t   Ae  1 n   n t 
1  a sA  Ae 1 n   n t  sA  generalsolution
1  a   n    n 
 sA  1 n   n t
  z0 
sA
e  definitesolution
   n    n
Since the solution is in terms of z, we should transform it back to k.

98
University of Gondar Lecture Notes
 Mathematical Economics

 sA  1 n   n t
k 1 a t    k 1 a 0 
sA
e  definitesolution
   n    n
 sA   1 n   n t
As t becomes larger and larger,  k 1 a 0  e approaches to zero.
   n 
1
 sA  1 a
So, k 1 a t    k t   
sA
 , the time path of capital stock is determined
  n     n  
by the values of a, s, δ and A. this leads to the conclusion that the time path of k
approaches to a constant (steady state value) and that steady state value is
1
 sA 1 a
k t     and the steady state value of percapita output is
   n  
a
 sA 1 a
y t   A  , because y  Ak a
   n  
a
 s  1 a
y t   A
1
1 a
 
   n  

6.5 Difference Equation

In the previous sections of our discussion, we take time as a continuous variable. In
such cases the change of time can be made infinitesimal. That is why we call it
differential equation. When time is considered as a discrete variable such that it takes
only integer value, the concept of derivatives can not be applied. In such cases the pattern
of change of variables and their functional values are described as differences rather than
differentials. An equation describing the discrete change of variable with respect to time
is called difference equation. This discrete time version of economic dynamics is also
called period analysis.
In this section we will discuss about the simplest form of difference equation i.e. first
order linear difference equation.
Definition: - a difference equation is an equation involving differences. It can be viewed
at least from 3 angles; as a sequence of numbers, as a discrete dynamical system and as
an iterated function.

99
University of Gondar Lecture Notes
 Mathematical Economics

1. Difference equation is a sequence of numbers generated recursively using a rule

to relate each number to the previous number in a sequence. Example: the
sequence s  1,1,2,3,5,8,13, 21 is generated using the rule yk  2  yk 1  yk where
k=1, 2, 3, … and the initial value y(0)=1
2. a difference equation is an iterated map in such a way that
yk 1  f  yk 
If we see the sequence in 1 above as an iterated function it can be represented as
follows. y0, f  y0, f  f  y0....... Where f  y0 is the 1st iterated function
of y 0  .
A linear difference equation can be defined as an equation that relates a variable yt to its
previous values linearly. The simplest one is a first-order scalar linear difference equation
taking the form,
yt = ayt−1 + B, t,
This is a first-order difference equation, since only first lag of its own values yt−1
appears in the equation. Here, we are interested in finding a solution of y in terms of its
current, past, or future values.

6.5.1 Solving Difference Equation

A solution of a difference equation is an expression or formula that makes the difference
equation true for all values of the integer variable t.
The method of finding solution to difference equation basically follows the same logic to
that of solving difference equation. In finding solution to differential equation, we are
interested in the time path of a given function. Although we use the concept of
derivatives and differentials in the solution procedure, the final answer is free from
derivatives and differentials. In the same way, our final answer in solving difference
equation should be free from any y term although we use the concept of differences in
finding solution. In solving differential equation we have intensively used the concept of
integration. However; this concept can’t be applied for the case at hand since time is
considered as a discrete quantity.

100
University of Gondar Lecture Notes
 Mathematical Economics

ITERATION METHOD
The nature of difference equation allows the solution to be calculated recursively. It is
easier to see the solution of difference equation through algebraic method.
Example; given yt  1  ayt    with initial value y(0), it is possible to determine the
solution using algebraic method as follows.
at t  0, y 0  y0 initial value

t  1  y 1  ay0  
t  2  y 2  ay1    aay0  b     a 2 y0  a  1
.......... .......... .......... .....
.......... .......... .......... ....

t  n  y n   a n y0  1  a  ...  a n 1  
However; the series in the bracket of the right hand equation is a closed geometric series
 n, for.a  1
n 1

taking the values  a  1  a ni

i 0  1  a , ifa  1

Therefore using this recursive approach, the solution to a difference

equation yt 1  ayt   b with initial value y(0) is

 y0   t , for' '  1
 y   , for' '  1 & t  odd


0
y t    y0 , for' '  1 & t  even
  1  at 
 a t y0     , if ' a '  1

  1 a 

General Method
The general method, as the name suggests, provides a systematic general framework for
solving First-order difference equations. Suppose you are given a general first order
difference equation of the form yt  ayt 1  
The solution to this difference equation consists of two parts: a part named the particular
solution denoted y p which is any solution to the first order difference equation given

above and a part named the homogenous solution denoted y c which solves the difference

equation yt  ayt 1 . The general solution is the sum of y p and y c .

101
University of Gondar Lecture Notes
 Mathematical Economics

Let's begin by thinking about the homogenous solution, i.e. the solution to yt  ayt 1 since

at each stage y is  times the previous value, the solution to this would be of the form
yc   t A.......... .......... .......... .......... .......... ......1
Where A is a constant. The `A' term is important because we have to consider all possible
solutions to yt  ayt 1
Now let's think of the particular solution, which can be ANY solution to the
equation yt  ayt 1   . Since any solution is allowed, we can, for example, find a


solution that sets all the y's equal. If we set y  ay   we get the solution y  .
1

Combining yt  y p  yc we get the general solution of the form yt   t A  .
1
If the initial condition is known (y0 is given), we can definitize the above solution by
 
determining the exact value of ‘A’. Setting t=0, we get y0  A   A  y0 
1 1
So, the definite solution of the difference equation will be of the form
   
yt   t  y0   
 1  1
 
yt   t y0  t
1 1
1   t 
yt   y0  
t

 1 

Note, the final answer is similar with the one we get using iteration method. However; the
general method is used only when   1 . In the case of   1 , the iteration method is

used.
Example: solve the difference equation yt  0.2 yt 1  4 where y0  10 using the general
method.
Solution; let us start with the solution of the homogeneous system yt  0.2 yt 1 . Using the

formula in equation (1) above, the solution is yt  0.2t A , where A is a constant.

The particular solution is the solution where all y’s are equal; i.e.,
y  0 .2 y  4

102
University of Gondar Lecture Notes
 Mathematical Economics

y5

Combining the two solutions, we get yt  y p  yc

yt  0.2t A  5
Since we know the value of y 0 , we can definitize the solution by substituting a specific

value to A. y0  A  5  A  y0  5

yt  0.2t  y0  5  5

yt  0.2t 5  5

6.6. DYNAMICS OF FIRST ORDER LINEAR DIFFERENCE

EQUATIONS
Steady State Value
Having solved for the time path, we can examine the dynamic behavior of a variable (i.e.
its behavior over time). We will examine the stability of the difference equation as well
as its steady state value.
The steady state value is a value at which the endogenous variable exhibits no dynamic
adjustment. In other words, given the difference equation yt  ayt 1   , a steady state

value y * exhibits the property that yt  yt*1

*

A difference equation can have zero, one or more (in the case of non-linear difference
equations) steady state values.
In th e above difference equation, the steady state value is therefore

y *  ay*    y * 
1

Stability

A difference equation is said to be stable if lim yt is a finite quantity. If lim yt

t  t 

doesn’t exist, then the difference equation is said to be unstable.

A stable difference equation will converge to a steady state value. For example, let us
examine the difference equation in the above example.
yt  0.2t 5  5

103
University of Gondar Lecture Notes
 Mathematical Economics

t 

lim yt  lim 0.2t 55
t 
  5 . Therefore the difference equation is stable.

1   t 
For general form yt  ayt 1   where a  0 , and the solution yt   t y0     , it
 1 

will be stable only if a  1, in which case the steady state value will be

lim yt = lim( t y0  1   )  

t

 1 
t  1
t 

The equation will be unstable if   1 &   1 because, in such cases

1 t  Will be undefined approaching either or  

lim( y0  
t
 )
 1 
t 

If   1 , the solution is yt  y0  t and it will be divergent unless   0

yt  y0 , t  even
If   1 , the solution is
yt   y0   , t  odd
In this case, the equation will oscillate between the two values as t becomes larger and

larger in value. The only case in which this system converges is for y0  . So, it will
2
not be stable. In general, we can identify 7 cases for the dynamic behaviors of the
sequence based on the values of a. this are summarized as follows.
Value of a
lim  t
t 
Dynamic behavior of the function

 1  Diverges from the equilibrium (unstable)

 1 1 Stable
0   1 0 Stable
 0 0 Stable
1    0 0 Alternate b/n positive and negative values but approach
to equilibrium) (stable)
  1 - Alternate between 1 and -1 for even and odd values of t
  1 - Alternate b/n positive and negative values but diverge

104
University of Gondar Lecture Notes
 Mathematical Economics

from equilibrium) (unstable)

6.7 Practical Application of Difference Equation in

Economics
The cobweb model
In some markets, particularly agricultural markets, supply cannot immediately expand to
meet increased demand. Crops have to be planted and grown and livestock takes time to
raise. Some manufactured products can also take a while to produce when orders
suddenly increase. The cobweb model takes into account this delayed response on the
supply side of a market by assuming that quantity supplied now (Qst) depends on the
ruling price in the previous time period (Pt−1), i.e. Qst= f(Pt−1) where the subscripts
denote the time period. Consumer demand for the same product (Qdt), however, is
assumed to depend on the current price, i.e. Qdt= f(Pt )
This is a reasonable picture of many agricultural markets. The quantity offered for sale
this year depends on what was planted at the start of the growing season, which in turn
depends on last year’s price. Consumers look at current prices, though, when deciding
what to buy.
The cobweb model also assumes that:
 the market is perfectly competitive
 Supply and demand are both linear schedules.
The assumptions of the cobweb model mean that the demand and supply functions can be
specified in the format
Qdt  a  bpt &
Qst  c  dpt 1

Where a, b, c and d are parameters specific to individual markets. Assuming that market
clears at each period, we can transform the above two equations as 1st order linear
difference equation.

105
University of Gondar Lecture Notes
 Mathematical Economics

Qdt  Qst
a  bpt  c  dpt 1
dpt 1  bpt  a  c
d ac
pt  pt 1 
b b

 d  ac
t

The general solution for this equation is pt  A    .

 b bd
ac ac
At t=0 p0  A   A  P0 
bd bd
Using this initial condition, we can definitize the above solution as

a  c  d  a  c ac
t

pt   p0      . But,  p.
 b  d  b  b  d bd

 
t
 d
pt  p0  p     p
 b
 
In the last equation, the p0  p term shows whether the time path is above or below the
t
 d
equilibrium value. On the other hand, the negative sign in the term    dictates that
 b
t
 d
the time path will be oscillatory. Since b & d>0,    will have positive and negative
 b
values for even and odd values of t. as far as the values of b & d are concerned, there will
be 3 distinct oscillations in the cobweb model. This are
1. When d  b , the oscillation will be explosive, and the dynamic value of the
price level will centrifugally go farther and farther from the intertemporal
equilibrium
2. When d  b , the oscillation will be uniform
3. When d  b , the oscillation will be damping, and the time path of the price level
will centripetally converge.

Answer key
106
University of Gondar Lecture Notes
 Mathematical Economics

Exercise 1.1
 x  5  2  5  3
1. At x  2 , the function is continuous b/c
x 1  2 1  3

To check the differentiability, let us check the existence of limit of the

difference quotient.
f x  h   f x  x  h   1  x  1  1
lim  lim
h 0  h h 0  h
f x  h   f x   x  h   5   x  5  1
lim  lim
h 0  h h 0  h
Since the limit of the difference quotient doesn’t exist, the function is not
differentiable at x=2

2. A. y  x 2 log3x

dy 1 x
 2 x log3x  x 2  2 x log3x 
dx x ln 3 ln 3
b. g  x   e5 x  4 
dg
 5e 5 x  4
dx
c. f x   2 x 3  15x 2  36x  4
df
 6 x 2  30 x  36
dx
d 2g
3. g x   4 x  x 
dg 4 1
 1  2  
dx 2 x dx x

107
University of Gondar Lecture Notes
 Mathematical Economics

Exercise 1.2
a.

3x  11 3x  1 a b
x 2
 x6
dx  2  
x  x6 x6 x5
 ax  5a  bx  6b  3x  1
a  b  3 & 5a  6b  1
19 14
a &b 
11 11

3x  11 19 14
x 2
 x6
dx  
11( x  6)
dx  
11( x  5)
dx

19
lnx  6  14 lnx  5
11 11
u 1
u  2 x  1, x 
b. x 2 x 1dx let 2

 5 3

u 1 du 1  u 2 u 2 
 x 2 x  1dx   2 u 2  2  5  3 
 

2.  20  4 x   20x  2 x
1
3
2 3
1
 2

 203  23  20  2   56

Exercise 2.1
1. Q  p  , where a and B are constants

dQ p p
d  .  P  1   
dP Q p

d logay  dy
,a  c  . x
 
2.
d logax
dx y ?

3. Given 1 x 3  1 x 2  2 x  1 , find the critical points and determine whether

9 6 3
they are maximum or minimum.

108
University of Gondar Lecture Notes
 Mathematical Economics

 x  x  2   0,  x  1 & 2
dy 1 2
dx 3
d2y 1 d2y
 2 x  1, atx  1,  1  max imum
dx 2 3 dx 2
d2y
x  2, 2  1  min imum
dx
4. C (Q)  aQ 2  bQ  c, (Q  0).

b
First order condition: C ' ( x )  2aQ  b  0  Q 
2a
Second order condition: C ' ' ( x )  2a . The cost function will have minimum
point only if the second derivative is positive. So C(x) will have minimum iff a>0.
5. C (Q)  aQ 3  bQ 2  cQ  d

C ' ( x )  3aQ 2  2bQ  c

b
C ' ' ( x )  6aQ  2b  0  Q   , inf lection
3a
b
C ' ' ( x )  0, Q  concaveup
3a
b
C ' ' ( x )  0, Q  , concavedown
3a

Exercise 2.2
1. Quadratic approximation of f x   5x  3 , about x=0
2

f 0  32 
1
9
f ' ( x )  105 x  3  f ' (0)  30
f ' ' ( x )  10
1
f ( x )   30x  10x 2
9
2. Quadratic approximation of F(K)=AK  about K = 1

109
University of Gondar Lecture Notes
 Mathematical Economics

F (1)  A
F ' (k )  aAk a 1  F (1)  aA
F ' ' ( k )  a a  1Ak a 2  F ' ' (1)  a (a  1) A
F (k )  A  aA(k  1)  a ( a  1) A(k  1) 2

3. Prove that 1  x   1  mx for x close to zero

m

We can easily come up with the proof using linear approximation of the left hand
function about x=0

f (0)  1  0  1,
m

f ' ( x )  m(1  x ) m1  f ' (0)  m

f ( x )  1  mx; proved

Exercise 2.3
MP2
1. X 1  ,
P12

a. if there is a neutral inflation it means that there is a proportional increase in

the price of the two commodities. The effect of such an increase in the prices
on the demand can be determined as follows. Let’s assume that there is a
proportional increase in the price of the two commodities by a rate h. the new
price will be Pn  (1  h ) P . Let r  1  h  Pn  rP

MrP2 MP2 x
2
 2
 r 1 x1  1 . So the new quantity demanded is the
rP1 rP1 r
quantity demanded before inflation net of inflation.
b. By Euler theorem if a function is homogeneous of degree r, the frist order
partials should be homogeneous of degree r-1.

x 2 MP2 P1  2 M ( rP2)(rP1)  2  2 MP2 P1 

  = r   
P1 P14
( rP1 ) 4  P14 

x M M M
 2   r  2  2  . So it satisfies Euler’s theorem
P2 P1 ( rP1 ) 2
 P1 

110
University of Gondar Lecture Notes
 Mathematical Economics

2. f  x   6  8 x  7 x 2  3x 3  2 x 4

Solution to the self test Exercises

Chapter – Three
1. Given
Cost: C x, y   0.04x 2  0.01xy  0.01y 2
Price of x: Px  15
Price of y Py  9
  TR  TC where  = profit
TR = Total Revenue
TC = Total Cost
TR  TRx  TR y
  15x  9 y  0.04x 2  0.01xy  0.01y 2  4 x  2 y  500

F.O Partials

 0 : 15  0.08x  0.0 y  4  01
x

 0 : 9  0.01x  0.02 y  2  02
y
From (1), we have 11 = 0.08x +0.01y or 0.08x +0.01y = 11 (3)
From (2), we have 0.01x +0.02y=7 (4)
Equations (3) & (4) can be solved by one of the following ways
- substitution method
- Inverse method
- Cramers rule
- Gaussian Elimination method
Let us now use the Cramers rule

Ax 0.08 0.01  x  11

X   0.01 0.02  y    7 
A      
A  0.080.02  0.010.01
 0.0016  0.0001  0.0015
11 0.01
Ax   110.02  70.01  0.22  0.07  0.15
7 0.02
0.15
Hence x   100
0.0015
Second Order Sufficient Condition

111
University of Gondar Lecture Notes
 Mathematical Economics

 xx  0.08  xy  0.01   yx ,  yy  0.02

 xx  xy
H
 yx  yy
 0.08  0.01 H 1  0.08  0

 0.01  0.02 H 2  H  0.0015  0
Ay
Y
A
0.08 11
Ay   0.087   0.0111
0.01 7
 0.56  0.11
 0.45
0.45
Therefore Y   300
0.0015
The value (amounts) if x and y that maximize the profit of the firm are 100 and 300
respectively.
2. Given production firm Q  f  L, K   6 k 3 L
PQ  0.5
Pk  0.1 R  0.5 / 6 K 3 L
3 1
PL  1 R  3k 2 L 2

c  0.1K  L
  3 k 3 L  0.1k  L
3 1
 3k 2 L 2  0.1k  L
F.O.C.
 3 3 2 1 2
 k L  1  0  1
L 2

 3. K 2 L 2  0.1  0  2 
3 1 1
k 2
3 32
k
from 1, 2 1 1
2
L
2 3 3 2  12  2
x k  L 
3 2  3
3 2 1
k 2  L2
3

112
University of Gondar Lecture Notes
 Mathematical Economics
2
3 X2
 2 1 3
k 2 3
 L 2
3 
2
2 3 1
k   L 3
3
1
4 3 1
  L 3
9
1
4  3
k   L   3
9 
Substitute the value of k form equation (3) into (2) to find the value of L.
1
 
1 2
3  4  3  12
3  L L  0 .1
2  9  
 
1
9  4  6 12
L L 0
2  9 

1
1
9  4 6 6 2
1

  L L  0.1
29
1
2
9  4 6 3
  L  0.1
29
CHAPTER 3
3, Solutions to the self test exercises
Given: constant unit price=P
Inputs of x1 and x2 are purchased at 4 constant unit factor prices of w1 and w2
respectively.
The firm faces competitive input and output market and the production function in given
by

113
University of Gondar Lecture Notes
 Mathematical Economics

Q  x1. x2 
Note that  MPx1 when MP  marginal product
x1
Q  x1. x2 
 MPx 2
x 2
PQ  x1. x2 
Thus  VMPx1
x1
PQ  x1. x2 
 VMPx2
x 2
Dividing equation (1) by (2) yields
PQ  x1. x2  x1 w1

PQ  x1. x2  x2 w2
VMPx1 w1

VMPx2 w2
This means that optimal factor utilization is achieved when the ratios of the value
marginal products of the respective factors are equal to the ratios of the input prices.
The second order condition requires that the slope of the VMP of the X’ s be lower than
of the stop of the input (factor) prices i.e.
Qx1.x2 
  PQ  C Q  Qx1.x2 
C  w1 x1  w2 x2
  PQ x1.x2   w1 x1  w2 x2
  x1.x2   PQ x1.x2   w1 x1  w2 x2
Taking the F.O.C
 PQ x1.x2 
0  w1  0 1
x1 x1
 PQx1.x2 
0  w2  0 2
x2 x2
Given   PQ - C
  x1 .x2   PQ  x1 .x2   w1 x1  w2 x2
F .O.C
  x1 .x2  PQx1 .x2 
 0,  w1  01
x1 x1
  x1 .x2  PQx1 .x2 
 0,  w2  02 
x2 x2

114
University of Gondar Lecture Notes
 Mathematical Economics

S .O.C
 2  x1 .x2  P 2  x1 .x2  w1
 0, 0 b/c 0
x1
2
x1
2
x1
 2  x1 .x2  P 2  x1 .x2  w
 0, 0 b/c 2  0
x2
2
x2
2
x2
4. Consider a monopolist firm
Given products Q1, Q2
The demands facing the monopolist firm are given as follows :
P1=55-Q1-Q2
P2=70-Q1-2Q2
C Q1.Q2   Q1  Q1Q2  Q2
2 2

PQ   TR1  TR2  TC

 P1Q1  P2Q2  TC

 55  Q1  Q2 Q1  70  Q1  2Q2 Q2  Q1  Q1Q2  Q2

2 2

 55Q1  Q1  Q1Q2  70Q2  Q1Q2  2Q2  Q1  Q1Q2  Q2

2 2 2 2

 55Q1  70Q2  3Q1Q2  2Q1  3Q2

2 2

F.O.C
 Q1 , Q2 
 0,55  3Q2  4Q1  0  1
Q1
 Q1 , Q2 
 0, 70  3Q1  6Q2  0  2
Q2
From equation 1, we have
4Q1  3Q2  55
3Q1  6Q2  70
4 3  Q1  55
A       is matrix form
3 6 Q2  70
Using Cramer's rule
AQ1
Q1 
A
A  64   33  15

55 3
AQ1   556   703
70 6
 330  320
 120

115
University of Gondar Lecture Notes
 Mathematical Economics

AQ1 120
Q1   8
A 15
4 55
AQ2   470  355  280  165  115
3 70
AQ2 115
Q2    7.67
A 15

S.O.C
 11  12
H 
 21  22
 11  4  21  3
 12  3  22  6
4 3
Thus, H 
3 6
H 1  4  0, H 2  H  4 c    3 3
 24  9
 15  0
Since the values of the Hessian determinant alternate in signs, profit is maximized at the
value of output where Q1  8 & Q2  7.67
5. The demands functions facing the monopolist are given by
P1  63  4Q1
P2  105  5Q2
P3  75  6Q3
The total cost function is C(Q)=20+15Q
Where Q = Q1+ Q2+ Q3
  TR  TC TR  TR1  TR2  TR3

Profit   is maximized when MR1=MC1

MR2=MC2
MR3=MC3

116
University of Gondar Lecture Notes
 Mathematical Economics

  TR1  TR2  TR3  TC

 P1Q1  P2 Q2  P3Q3  TC
 63  4Q1 Q1  105  5Q2 Q2  75  6Q3 Q3  20  15Q1  Q2  Q3 
because Q  Q1  Q2  Q3
 63Q1  4Q  105Q2  5Q  75Q3  6Q  20  15Q1  15Q2  15Q3
1
2 2
2
2
3

 48Q1  4Q12  90Q2  5Q22  60Q3  6Q32  20

F.O.C

 48  8Q1  01 Q1  6
Q1

 90  10Q2  02  Q2  9
Q2

 60  12Q3  03 Q3  5
Q3
S.O.C
 11  12  13
H   21  22  23
 31  32  33
8 0 0
 0  10 0
0 0  12
H 1  8  0
H 2  80  0
H 3   960  0
Thus, the monopolist maximizes profit when it produce Q1=6, Q2=9 and Q3=5 and total
Q =Q1+Q2+Q3
= 6+9+5
=20
P1  63  4Q1 P2  105  5Q2  P3  75  6Q3
 63  46 105  59  75  65
 69  24  105  45  75  30
 39  60  45
6. P1  100  Q1
P2  80  Q2
c  6Q Q  Q1  Q2

117
University of Gondar Lecture Notes
 Mathematical Economics

  TR1  TR2  TC
 P1Q1  P2 Q2  c
 100  Q1 Q1  80  Q2 Q2  6Q1  Q2  becasue Q  Q1  Q2
100Q1  Q12  80Q2  Q22  6Q1  6Q2
  94Q1  Q12  74Q2  Q22
F.O.C

 94  2Q1  0 Q1  47
Q1

 74  2Q2  0, Q2  37
Q2
i.

a ) Q  Q1  Q2
Q1  47 and Q2  37 would be sold in the
 47  37
the two markets.
 84
b) P1  100  Q1 P2  80  Q2
 100  47  80  37
 53  43
c)   94Q1  Q12  74Q2  Q22
 9447  47  7437  37
2 2

ii) If price discrimination is made illegal

a) In the absence of price discrimination P1 = P2= P

Thus, P1  100  Q1  P  100  Q1

100 – P =
Q1 P2  80  Q2  P  80  Q2

80 – P = Q2
Beside Q = Q1 +Q2
= 100- P +80 –P
Q = 180 =2P
1
P =90- Q
2
Therefore,

118
University of Gondar Lecture Notes
 Mathematical Economics

  R C R  PQ
 1 
  90  Q Q  6Q
 2 
1
 90Q  Q 2  6Q
2
1
  84Q  Q 2
2
F.O.C.

 84  Q  0
Q
Q  84
1
b) Since P  90 - Q
2
 90 
1
84
2
 90  42
 48
c)   R  C
1
 840  Q 2
2
 8484  84
1 2

2
 7056  7056
1
2
 7056  3528
 3528
This is the amount of profit that the monopolist gets when it does not practice price
discrimination.
The amount of profit loss= profit obtainable with price discrimination- profit obtainable
without price decimation.

119
University of Gondar Lecture Notes
 Mathematical Economics

Exercise 4.1
1. y  3x  5 x1  x1 x2  6 x  4 x2  2 x2 x3  4 x32  2 x3  3x1 x3,
2
1
2
2

First order condition

y
 6 x1  5  x 2  3x 3  0  6 x1  x 2  3x 3  5
x1
y
  x1  6 x 2  2 x 3  4  0   x1  6 x 2  2 x 3  4
x 2
y
 3x1  2 x 2  8 x 3  2  0  3x1  2 x 2  8 x 3  2
x 3
6  1  3 x1 5
1 6 2 x2  4
3 2 8 x3  1
Using crammers’ rule
5 1  3 6 5 3 6 1 5
4 6 2 1 4 2 1 6 4
1 2 8  3 1 8  3 2 1
x1  , x 2  , x 3  
6 1  3 6 1  3 6 1  3
1 6 2 1 6 2 1 6 2
3 2 8 3 2 8 3 2 8
Second order condition ;
f 11  6, f 12  1, f 13  3, f 21  1, f 22  6, f 23  2, f 31  3, f 32  2, f 33  8

6 1  3
So,the third order hessian matrix will be H   1 6 2 .
3 2 8
6 1
Principal minors H1  6  0, H 2   5  0, H 3  214  0 so, the point is
1 6
maximum.
3. Maximize c  45x 2  90xy  90 y 2 , s.t 2 x  3 y  60,

L  45x 2  90xy  90 y 2   60  2 x  3 y 

l
 90x  90 y  2  0
x
l
 90x  180 y  3  0
y
l
 2 x  3 y  60


120
University of Gondar Lecture Notes
 Mathematical Economics

0 45 1 45 0 1
0 60 1 30 0 1
45 45  1 x 0
60 3 0 2 60 0
30 60  1 y  0 , x   6.66, y  
45 45 1 45 45 1
2 3 0  60
30 60 1 30 60 1
2 3 0 2 3 0

45 45 0
30 60 0
2 3 60
 
45 45  1
30 60  1
2 3 0
90 90 2
90 90
s.o.c H H  90 180 3  H 1  90  0, H 2   0, H 3  0 so,the
90 180
2 3 0
critical values are maximum points.
5. Maximize u x, y   x 0.25 y 0.4 , subjectto, 2 x  8 y  104
l  x 0.25 y 0.4   104  2 x  8 y 
l
 0.25x 0.75 y 0.4  2  0
x
l
 0.4 x 0.25 y 0.6  8  0
y
l
 104  2 x  8 y  0

x  4.5, y  12

121
University of Gondar Lecture Notes
 Mathematical Economics

References
Chaing A.C. (1984), Fundamental Methods of Mathematical economics, 3rd ed. McGraw
Hill Inc.
Dowling, E.T. (1980), Introduction to Mathematical Economics, 2nd ed. Schuam’s outline
series, McGraw-Hill
Kapoor V.K. (2002), Introductory Mathematics for Business and Economics, Sultan
Chand and Sons
Klein M. (2003), Mathematical Methods for Economists
Mike Rosser (2003), Basic Mathematics for Economists, 3rd ed. Routledge.
Sydsaeter, K. and Hammond, P. (2002), Essential Mathematics for Economic Analysis,
Prentice Hall
Sydsaeter, K. and Hammond, P. (2000), Economists’ Mathematical Manual, 3rd ed.
Yamane, T. (1978), Mathematics for Economists, An Elementary Survey. 2nd ed.

122
University of Gondar Lecture Notes