Session 10 – Introduction to

Difference Equations

Slide 1
 Session Overview

• Identify the components of the general solution of a

difference equation
• Cross check you answer to general solution of a first
order linear difference equation
• Identify the components of the general solution of a
difference equation
• Discuss the time path of a solution to difference
equation
Slide 2
 Session Outline
The key topics to be covered in the session are as follows:

• Definition and Types of Difference Equations

• Solving Difference Equations (Iterative
Methods)
• General Formula for first order difference
equations
• Dynamic Stability Conditions of Equilibrium

Slide 3
 Reading List
• Dowling, E. T. Introduction to Mathematical Economics
(Second Edition), Schaum’s Outline Series, McGRAW-
HILL. ISBN 0-07-017674-4
• A. C., Chiang. (1984) Fundamental Method of
Mathematical Economics, McGraw Hill international
• Raymond A. B., Ziegler, M. R, and Byleen K. E. (2000).
Applied Mathematic for Business, Economics, Life
Sciences and Social Sciences. (Seventh edition).Prentice
Hall. ISBN 0-13-083120-4.

Slide 4
 Topic One
Definition and Types of Difference Equations

• A difference equation is any equation that expresses a

relationship between a dependent variable and a lagged
independent variable which changes at a discrete interval
of time. Example:
It = f(Yt-1)
• where It is investment in the years t and Yt-1 is income in
the previous year t-1 .
• Investment in current year t - It is a function of previous
years t-1 income or income lagged by one-period (Yt-1).

Slide 5
 Examples of a Difference Equation
• Current period savings - St depends on a function of
previous year’s income -Yt-1

• Current period investment - It depends on or is a function

of previous period’s income - Yt-1

Slide 6
 Identifying Difference Equations
• Generally, we identify a difference equation by
the maximum time-lagged.
• First order difference equation refers to one in
which the maximum time lag is one period, i.e.

• Second order difference equation refers to a

difference equation in which the maximum time
lag is two periods, i.e.

Slide 7
 Types of Difference Equations
• We classify difference equations as either homogenous or
non-homogenous. With this classification, one can combine
the order and homogeneity in naming a difference equation.
• Consider an nth order difference equation of the form:

• If C=0, then the difference equation is referred to as a

homogenous difference equation.
• If C0, then the difference equation is said to be non-
homogenous.
Slide 8
 Types of Difference Equations cont’d

- First order homogenous

difference equation because the maximum time
lag is one t-1 and there is no constant.

- Second order non-homogenous

equation because the maximum time lag two t+1
and there is a constant -3

Slide 9
 Activity 1

1. What is a difference equation?

2. Give two examples of a difference equation.
3. Using the order and the homogeneity, name the following
difference equations:

4. Provide 3 real example of economic situations that can be

expressed in difference equations

Slide 10
 Topic 2
Solving Difference Equations (Iterative method)
• The solution to a difference equation defines y for
every value of t and does not contain a difference
expression

• 1. Iterative method
• 2. General Solution of first order Difference
Equation

Slide 11
 The Iterative Method

• Simple method that permits us to write down the general

solution of a difference equation after careful observation
of how the equation behaves at successive periods.

• Requires that we successively substitute the values of t =

0, 1, 2, 3 … into the given difference equation and from
the observation of the successful values generate how the
equation generally behave. Using the iterative method
requires that you obtain information on the value of
the first term, for instance yo.
Slide 12
 Example: Iterative Method
• Consider the first order difference equation
• Find the general solution using the iterative method if the initial value is y0
• SOLUTION
Substitute the values t = 0, 1, 2 into equation (2) and out of the observation
deduce the general solution.
………………………………………(1)

t=0, implies ---------------------------(2)

t=1, implies ---------------------------(3)
from equation (1) when t=0, hence inserting this value into equation
(3) gives:
---------------------------(4)
Slide 13
 Example: Iterative Method cont’d
• From equation (1) when t=2, the difference equation gives
……………………………………….(5)

• From equation (2) when t=1, gives and inserting this

value into equation (4) and simplifying it gives:

• Analogically if t=3, we obtain:

Slide 14
 Example Using the Iterative Method
• Notice that when
t=0, t=2,

t=1, t=3,

• Hence, we see that by substituting successively the values t = 0, 1,

2, 3… we can for any time period, t, express the general solution as:

• Notice that the initial value is constant, and thus c plays a crucial
role in determining the values of y as t changes.

Slide 15
 Example Using the Iterative Method
• Consider the difference equation
• Find the general solution using the iterative method if .

SOLUTION

• This example is the same as the previous one. The only difference
is that the constant c is now a specific figure - [3].

• To solve this problem using the Iterative method, substitute

successive values of t = 0, 1, 2, 3… and find the general formula.

Slide 16
 Example Using the Iterative Method
• When t = 0, or

• When t=1, or since

• When t=2, or since

• The general solution can therefore be deduced as

Slide 17
 Crosschecking your answer

• This involves crosschecking if any successive values

satisfy the given equation.
• In the previous examples, lets crosscheck if our answer is
correct or not
• Solution
Recall that the equation is and the general solution
is:
The method required that we (1) take any successive real
values, (2) insert it into the solution and (3) ascertain if it
satisfies the equation.
Slide 18
 Crosschecking your answer
• We will take the value t=0 and t=1
If t=0,
If t=1,
Our difference equation is :
Putting these values into the difference equation we obtain:
or

The two solutions of t=0 and t=1 satisfies the equation.

Slide 19
 Topic 3: General Formula for First Order Linear
Difference Equations

• Consider a first order linear difference equation of the

form:

• where b and a are constants. Recall that first order

implies that the maximum time lag is one period.
• This equation is linear because all the variables are
raised to a maximum power of one and contain no cross
product (i.e., a multiple of two or more variables).
Notice that the variable is the dependent variable,
while the lagged variable is the independent variable.
Slide 20
 General Formula for First Order Linear
Difference Equations cont’d
• The general formula for the solution to the difference
equation is given as:
• if

• OR if b=1
Notice that the is the initial condition and often given in
a problem.
If we assume that and ,

Slide 21
 General Formula for First Order Linear
Difference Equations
• Then the general solution can be expressed as:
……………….. (1)
• If the value of the initial condition is given, then
the solution is called a definite solution.
• If it is not given, then we usually use an arbitral
constant

• In such a situation we call the solution a general

solution
Slide 22
 Example
• Solve the difference equation using the
general formula if .

• Crosscheck your answer.

Slide 23
 Solution
a. Recall that the general formular for the first order
linear difference equation is:
• ………………..(1)

• Re-writing our equation in the same form we obtain:

…………….......(2)
• Comparing equations (1) and (2), you can observe
that:

Slide 24
 Solution cont’d
• Notice that z0 =4. Thus, we have all the inputs to obtain the general
solution.
• We substitute the values of a, b and z0 into the general
solution formula, since

and obtain

• Simplifying this equation we obtain:

Slide 25
 Solution
b. Now to crosscheck our answer, we will have to
substitute any two successive values of t = 0, 1, 2, 3…
and see if what we will obtain satisfies the equation.
For simplicity, we will use t=0 and t=1. Notice that our
general solution is:

• If t=0,

• If t=1,
Slide 26
 Solution
• These two successive values of z should satisfy the original
difference equation
…………………..(3)
• Substituting the values of z0 =4 and z1 = -9 into equation (3) and
simplifying makes the LHS of the equation equal the RHS.
Specifically:

• Hence our general solution is correct.

Slide 27
 Activity 1
1. If investment is given as , where . Find
the general solution using the iterative method.
2. Using the iterative method, find the general solution to
the difference equation: .
3. Solve the following using the iterative method

4. Given that and , find the general

solution.
Slide 28
 Topic 4:
Dynamic Stability Conditions of Equilibrium
• Dynamic stability talks about how the solution of a first
order defence equation behaves as time increases beyond
bounds .
• Recall that the general solution of a first order linear
difference equation can be written as:
where and for

• The solution is made up of two additives and c.

• The expression is referred to as the complementary
function and the value c is referred to as a particular
solution. Slide 29
 Dynamic Stability Conditions of Equilibrium
cont’d
• Consider the general solution::
• This equation is dynamically stable if the complementary
function approaches zero as time -t approaches infinity.
• Mathematically:

• Now under what condition will this happen?

• Notice that the values A and b are constants

Slide 30
 Dynamic Stability Conditions of Equilibrium
cont’d

• The only variable in the expression will be time, t.

• Since t is expressed as a power of b whether or not
the function will approach zero as t approaches
infinity, will depend largely on the value of b.
• Consider the general solution. ,
• Assume that A=1, c = 0
• so that,
• Lets plot this function for different values of b

Slide 31
((1) 2102 
t 1)
t t tt t t
b (2)
(0)
b1/ 

2)
1/

Dynamic Stability Conditions of Equilibrium

cont’d

Value of b Absolute Value of b t t = 0 t = 1 t = 2 t = 3 t = 4

terms

b > 1 b 1 (2) t 1 2 4 8 16

b = 1 b 1 (1)t 1 1 1 1 1

0 < b < 1 b 1 1 / 2 
t 1 1/2 1/4 1/8 1/6

b = 0 b 0 (0) t 0 0 0 0 0

-1 < b < 0 b 1  1 / 2 
t 1 -1/2 1/4 -1/8 1/16

b = -1 b 1 (  1)t 1 -1 1 -1 1

b < -1 b 1 (  2)t 1 -2 4 -8 16

Slide 32
 Dynamic Stability Conditions of Equilibrium,
cont’d
a) When b assumes a value greater than 1 (i.e. b > 1),
for instance if b = 2, we observe the time path
(graph) for the function is:

• The time path approaches infinity as time approaches

infinity or it diverges
Slide 33
 Dynamic Stability Conditions of
Equilibrium
b) When b assumes a value equal to 1 (i.e. b = 1), the time
path is constant as follows:

The time path is constant as time approaches infinity or

neither converges nor diverges.
Slide 34
 Dynamic Stability Conditions of Equilibrium

c) When b is between zero and 1 (i.e. 0 < b < 1), for

instance, b = ½, then the time path will be :

The time path tends to zero as the time approaches zero or

converges to equilibrium (zero) over time.
Slide 35
 Dynamic Stability Conditions of Equilibrium,
cont’d

• When b = 0, the time path for will like this

Slide 36
 Dynamic Stability Conditions of Equilibrium
cont’d

• When b is between negative one and zero (i.e. -1 < b <

0), for instance where b = -1/2, then the time path for
will look like this:

Time path decreasingly oscillate about the equilibrium

y=0 and converges to the equilibrium as t approached
infinity. Slide 37
 Dynamic Stability Conditions of Equilibr
cont’dium

d) When b = -1, then we have the following time path for

Time path oscillates uniformly about the equilibrium y=0

between positive one and negative one. It neither converges
nor diverges.
Slide 38
 Dynamic Stability Conditions of Equilibrium cont’d

e) The final scenario involves the case where b is less than

negative 1 (i.e. b < -1). Specifically I will take b=2 and
obtain the following equation . The time path
of this equation will look like this:

Time path here oscillates about the equilibrium y=0 and

diverges, or moves away from the inter-temporal
equilibrium as time turns to infinity.
Slide 39
 Dynamic Stability Conditions of Equilibrium
cont’d
Time path diverges or converges and
Time path is oscillating or non-oscillating.

• If |b| > 1 then the time path diverges away from the equilibrium.
• If |b| < 1 then the time path converges towards to the equilibrium.
• If b > 0 then the time path is non-oscillating about the
equilibrium.
• If b < 0 then the time path will oscillate about the equilibrium.
• We use these two principle convergence/divergence and
oscillatory/non-oscillatory to describe the dynamic stability of a
solution to a first order linear difference equation.
Slide 40
 Dynamic Stability Conditions of Equilibrium

• Now let see what the constant A and c can do to the general
solution,

• A represents a multiplicative constant. It does not affect the

fundamental behaviour/ pattern of movement of the time path,
but rather scales it up or down depending on the magnitude .
• -If A>1 then it scales up the value of
• - If 0<A<1, then it scales down the value of
• The constant c represents the intercept,
• If , the vertical intercept of the graph is affected.
• The graph shifts up or down as the value of c increases or
decreases. Slide 41
 Example
Consider the difference equation
i. Find the general solution
ii. Crosscheck your answer
iii. Examine the dynamic stability of the solution

Slide 42
 Solution
i. Let us begin by re-writing the problem in the general
form by dividing through the equation by 4 as:

• Comparing equation (1) with the general form, I observe

that, a=0, b=4/3.
• The general solution can be written as:

, since

Slide 43
 Solution cont’d
Inserting the value of a and b in the equation we obtain

or

…………………………………(2)

Equation (2) is therefore the general solution.

Slide 44
 Solution cont’d
ii. Cross checking our answer
• We substitute the values t = 0, 1, into the general solution
(2);
• At t = 0,
• At t = 1,
Putting these values into the difference equation we obtain:

Slide 45
 Solution
• Since we have the values of Y1 and Y0 from the general
solution, we can substitute these values into (3) to verify
if the RHS = LHS;

• Hence since the RHS = LHS, our general solution is

correct.

Slide 46
 Solution
iii. Dynamic Stability
Recall that my general solution is given as:
The complementary function is and the particular
solutions, c = 0.
From the complementary function, b = 4/3.

Since =4/3 >1, the time path is non-oscillating, and also, because
=4/3 >0, the time path will not converge over time.

The time path, therefore is not dynamically stable and will not turn to
zero over time. In other words as t increases beyond bounds or
approaches infinity the time path will not converge to zero.
Slide 47
 Activity

1. Under what condition will a time path be said to be dynamically

stable?
2. Find the general solution of the following difference equations,
check your answer when t = 0, & t = 1, and comment on the
stability on the time path;
a. and
b.
c.
3. A population model for a population, , in year t, is given by the
difference equation; =50. Given that , find out if the population
converges over time.
Slide 48