TOPIC 1: INTRODUCTION TO ECONOMIC MODELING AND COMPUTER

SIMULATION
Introduction

Economists often seek to understand the nature and functioning of various types of economic

systems. In doing so, they rely on the relationships that exist among various economic variables,

and use such relationships to make predictions or forecasts of how the system will change, if

economic agents change their behavior or plans, or if variables adjust.

This process, however, must be guided by economic theory in order to establish such

relationships (models), to test their predictability or forecasting power, and if possible, revise

such models in the light of new information. This is the concept of economic modeling, which

simply put, is the process of creating models to study some phenomena of interest.

Economic modeling is generally defined as ‘a scientific approach of solving economic problems

through creation of appropriate models, being guided by economic theory, and utilising relevant

data and the appropriate analytical requirement, i.e., skills and tools.’

Economic modeling can be done either at the macro- level that involves modeling an aspect for a

larger aggregate, such as models for a whole country or region, or they could also be done at the

micro- level that involves modeling a particular aspect of interest relevant to the analysts.

Today, economic modeling has evolved and has found a significant part in decision making, both

in the public and private sectors, academia and even in international organizations. Ministries,

Central Banks, parastatals, Research organizations, University departments, international

organizations such as the IMF, African Development Bank, the World Bank and many other

institutions are using models in making decisions.

For this reason, computers have become essential for economic modeling. Various computer

software have been developed in this regard, for commercial purposes, including EVIEWS,

STATA, SPSS, PCGIVE, LINDO, LIMDEP, LISREL, PASCAL, SHAZAM, MINITAB,

GAMS, DEA, GRETL, among many others. However, some software are specialized for

particular modeling exercises only, while others are specialized for simulation. Computers will

remain an invaluable part to the applied econometrician or modeler, both in their training and

professional work. Economic modeling and computer simulation is therefore at the heart of

empirical economics and research, and is rapidly expanding field of study.

Objectives

Objectives by the end of this topic you should be able to:

1. Define the key terms used in Modeling.
2. Discuss the features of a model
3. Describe the elements of an economic model.
Learning Activity 1.1: Reading
Read the provided topic notes on Introduction to Economic Modeling and Computer Simulation.
Learning Activity 1.2: Journal

In not more than 500 words summarize the features of a model. Post your contribution in the
provided journal.
Learning Activity 1.3: Assignment

Distinguish between endogenous and exogenous variables.

Assessment

The journal in activity 1.2. and Assignment in activity 1.3 will be graded
Topic Resources
URL link

https://www.youtube.com/watch?v=qiQhAyuH2P4

Please go through the video link above to attempt the assignment above.
 Topic 1 Notes

Economic Models

Definition of a model

A model refers to anything that is used to represent reality, or the working of a particular system

of interest. Thus, a model is an abstraction from reality.

The reason why we abstract from reality is due to the fact that the real world has a lot of

complexities, interactions and surprises, and thus, we cannot claim to be able to study the

complex and intermediate nature of the real world or economy in totality.

Instead, we seek to capture only the salient or most important features of the system (the

essentials) of the real situation which it represents that are of particular interest to us, and then

make an assumption that the other relevant factors will remain constant for the period under

review, ie., the ceteris paribus assumption.

Another reason why we abstract from reality is due to the recognition that not all the inter-

relationships in the variables of interest are of equal importance in the understanding of the

particular economic phenomena under study. Thus, we only focus attention on those

relationships that are relevant to the problem at hand.

Therefore, models do not describe the true economic world, but instead, they abstract from the

truth or reality. A model can be constructed at different levels of aggregation and sophistication.

Hence, the degree of abstraction depends on the PURPOSE for which the model is constructed,
the EXPERIENCE of the MODELER in terms of their knowledge of available techniques and

their expertise in constructing and implementing models, and the NATURE of the problem that

calls for the modeling.

However, abstraction from reality does not imply unrealism! It implies the simplification of

reality and the beginning of the understanding of the great complexity of the real world.

1.2.2 Features of a model

Any model has to satisfy two key features, i.e., simplicity and realism.

1. SIMPLICITY

- A model is a simplified representation of reality.

- Thus, a model should be similar to, but simpler than the real system it represents, i.e., a

close approximation to the system.

- Therefore, a model should not be so complex that it is impossible to understand and

experiment with it.

2. REALISM

- Despite the fact a model simplifies a real system, this does not justify unrealism.

- The model should in fact as much as possible represent the system.

- Thus, the modeler still has to incorporate some reasoning and judgement in developing

and implementing models, and not view the model as a perfect substitute for reasoning.

- Realism thus implies that a model may be relevant one situation (region or time), and be

irrelevant in another situation, mainly due to change in fundamentals or information.

In overall, a good model is actually a judicious trade-off between realism and simplicity, through

logical abstraction.

Since both simplicity and realism are vital features for a model, the recommended practice in

modeling is to increase the complexity of a model iteratively, based on the need, the situation

(region or time), and the purpose; all aimed at enhancing model realism.

In summary, a model needs to fulfill the following critical features:

(i) The model is relatively simple and easy to understand and use,

(ii) The model is guided by the relevant economic theory in its development,

(iii) The model aspires to as much as possible represent the system – and this can be done

iteratively through computer simulations,

(iv) The assumptions behind the model are specified, and should be valid as well,

(v) The model outlines its structure, i.e., variables used and types of equations.

(vi) Any model is time- and location- bound – any model can only be true at a particular

time and location since model features may change due to change in fundamentals

and information, and these depend on time and region.

1.2.3 Model validity

Model validity is an important issue in modeling. It involves comparing model output with

system output. Thus, the ease and closeness with which a model reflects the real system is a key

component for a valid model.

By model validity, were interested in the significance of the variables used in the model, the

validity of the model itself to effectively address the problem being examined, and also the
validity of the estimation techniques, whether Ordinary Least Squares (OLS), Generalized Least

Squares (GLS), Simultaneous Equations, Maximum Likelihood Method (MLE), Generalized

Method of Moments (GMM), and so on.

The validity of a model may be judged on several criteria:

(i) Its predictive or forecasting power or performance,

(ii) The consistency and realism of its assumptions,

(iii) The extent of information it provides, - a valid model should provide as much

information as possible, but this depends on the purpose or use to which the model is

put.

(iv) Its generality – that is, the number or range of cases to which it applies,

(v) Its simplicity. A model that is simple can be replicated elsewhere and can have a long

life compared to a very complex model.

There is no general agreement regarding which of the above criteria is more important.

Economists thus hold differing views as follows:

- Milton Friedman proposes that predictive performance of a model is its most important

criterion for model validity.

Predictive performance is especially important when the purpose of the model is

forecasting the effects of a change in a certain variable.

- Paul Samuelson is of the view that realism of assumptions and power of the model in

explaining the behavior of the system or agents in the model is the most important

attribute to a model.
 Realism of assumptions and explanatory power are important attributes of a model if the

purpose of the model is the explanation of why a system behaves as it does.

- Most economists have the dominant position that the generality and simplicity of the

model are first-hand criteria: i.e., that a model’s validity is determined by the purpose or

use to which it is put, should provide as much information as possible, and should at least

represent the system.

Elements of an economic model

Although models are diverse, and therefore have varied elements, any typical economic model

should have at least the following 5 elements:

i. Endogenous variable

ii. Exogenous variables

iii. Model parameters

iv. Assumptions of the model

v. Closure rule

These are now discussed as follows:

(i) THE ENDOGENOUS VARIABLE:

- The endogenous (dependent) variable is the one whose value is determined within the

model. Thus, the value of the endogenous variable is determined by the other variables in

the model, and also the model parameters.

- For example, the famous Keynesian national income model is defined as:

Y = C + I + G + NX …………………………………. (1)

C = b0 + b1Y + b2Ct-1 + b3r ……………………..… (2),

 Where Y is national income or GDP, C is aggregate consumption expenditure, I is real

investment expenditure, G is the level of government expenditure, NX is net exports, C t-1

is income lagged by one period and r is the real interest rate. Also

In equation (1), Y is determined by the values taken by C, I, G, and NX. Effectively

therefore, Y is the endogenous variable or the regressand.

In equation (2), C is determined by the values taken by Y, C t-1 and r. Thus, C is the

endogenous variable.

- A model could have one or several endogenous variables, and this determines the level of

sophistication or detail that a model wants to have. Thus, a model is either a single-

equation model or a multiple-equation model, but this will depend on the purpose for

which the model is constructed. Thus, any economic model must clearly specify its

endogenous variable(s)

(ii) THE EXOGENOUS VARIABLE:

- An exogenous variable is one whose value is determined from outside the model, and

hence their values are defined elsewhere in other models, and are thus taken as given.

- However, it is important to recognize that a variable could be endogenous in one

particular model, but be an explanatory variable or regressor in another model. In

equation (1) above, C, I, G and NX are the exogenous variables. However, whereas C is

exogenous in equation (1), it is in fact endogenous in equation (2). Similarly, while Y

was endogenous in the first equation, it is now exogenous in the second equation, and so

on.

(iii) PARAMETERS
- A parameter is a scalar or constant which is either attached to a variable or it could be

stand-alone. Thus, there are two types of parameters: coefficient and intercept.

- Coefficient: - this is a scalar attached to a variable and it measures the partial derivative

of the endogenous variable with respect to the respective explanatory variables, holding

all other factors constant.

For example, in the equation: C = b0 + b1Y + b2Ct-1 + b3r, the scalars b0, b1, b2 and b3 are

the model coefficients. Thus, b1 measures the partial derivative of the dependent variable

C with respect to income Y, b2 measures the partial derivative of the dependent variable

C with respect to income Ct-1, and so on. Partial derivative is also referred to as Marginal

effect.

- Parameters are generally useful in a model since they define for us elasticities and

propensities. Thus, from equation 2 above, b1 is the marginal propensity to consume. If

the function is a log-log function, then the coefficients are readily defined as elasticities.

Elasticities and propensities are useful for policy purposes.

- Intercept, on the other hand, is stand-alone. It measures the expected (mean) value of the

endogenous variable if all other factors are held constant, or the point where the function

crosses the vertical intercept. Thus, b0 in equation 2 is the intercept, and it measures the

level of consumption when all other factors are held constant – i.e., autonomous

consumption.

(iv) ASSUMPTIONS

- Assumptions are statements or expressions that must be true or valid for a model to be

reliable and predictable. However, most importantly, it is not the assumption, but the

soundness and plausibility of the assumption, as informed by economic theory.

 - For example, in the theory of demand, the ceteris paribus assumption must hold for the

law to be valid or true. In the consumption function, the assumption made is that the

value of the marginal propensity to consume (MPC) lies between 0 and 1, i.e., 0<MPC<1.

- Recall that Paul Samuelson holds in esteem the consistency and realism of assumptions

as the key features for model validity.

(v) SOLUTIONS AND THE CLOSURE RULE

- Models are created not just for their own sake, but rather to help in forecasting and policy

analysis. To this end, any model must be capable of yielding solution values for the

specific variable(s) of interest.

- On the other hand, a closure rule helps to examine how equilibrium relationships are

obtained, or how a model will reach equilibrium, after a shock to the system.

- For most models, the closure rule usually consists of an identity, such as the national

income identity.

Characteristics of a Model

Models are diverse in terms of their features and specifications, but we can identify three key

characteristics for any economic model:

(i) Technical functions

(ii) Behavioral functions

(iii) Identity function(s)

TECHNICAL FUNCTION is one which describes technological or institutional relationships.

For example, the production function describes the relationship between output and inputs, and it

could be linear, cob-Douglas, Leontief, constant-elasticity of substitution (CES), and so on. The
tax function is also a technical function since it shows the relationship between tax revenue and

the determinants of tax, including some institutional factors, such as reforms in administration,

technology adoption, etc.

Technology may be labor-augmenting, capital augmenting or neutral (Total factor productivity).

BEHAVIOURAL FUNCTIONS specify the manner in which different groups in an economy

will behave in response to changes in various explanatory variables or factors. For example, the

consumption function shows how consumer expenditure on consumption of goods and services

as a function of disposable income. Also, the investment function is an example of a behavioral

function as it shows how the level of investment relates to the real interest rate.

IDENTITY is an equation stating the equality that holds true by virtue of the definition of the

variables involved. For example, the national income identity states as follows: Y º C + I + G +

X – M. thus, the left hand side has exactly the same meaning as the right hand side, or simply

put: Income is equal to expenditure. An identity helps to determine the closure rule of the

equilibrium position.

Types of Economic Models

Economic models can be classified based on: the time period, the number of variables or

equations, the purpose of the model, the type of data used in the model, and the methodology or

technique used to estimate the model.

Based upon these criteria, we can identify the following types of models: (i) Graphical or visual

models, (ii) Mathematical Models, (iii) Empirical Models, and (iv) Simulation Models.
(i) Graphical or visual models

- These are models that depict an economy or economic system by making use of graphs or

networks.

- A classic example of a graphical model is the aggregate demand-aggregate supply (AD-

AS) model, as shown below.

- Consider the following data on price (P), aggregate demand before (AD 0) , aggregate

demand after (AD1), and aggregate supply (AS) of a particular good in a given economy:

AD0 95 85 75 65 55 45 35 25

AD1 105 95 85 75 65 55 45 35

Qs 15 25 35 45 55 65 75 85

P 1 2 3 4 5 6 7 8

- Enter this information into excel spreadsheet but place it column wise.

- Now, highlight AD0, AD1 and AS then go to menu, and select: INSERT/ LINE and click

on the first graph that is indicated. You will notice the graph for AD-AS Model is

presented. Draw this diagram in the space given here below:

- The AD-AS model illustrates the effects of changes in either aggregate demand or

aggregate supply on the equilibrium price and quantity. The diagram above demonstrates

that an increase in aggregate demand will lead to demand-pull inflation.

- Apart from the AD-AS model, other examples of graphical models are: the IS-LM model,

the Phillip’s Curve, and the Mundell-Flemming Model.

- Although graphical models do not require heavy knowledge of mathematics, they are

actually based on mathematical principles and theorems.

- Graphical models have the advantage of being easier to understand compared to other

models, but they suffer from the 2- or 3- dimensional problem and thus may not capture

the complex nature of the real system.

(ii) Mathematical models

- Mathematical equations usually represent a system of equations with many variables. As

a result, solving these models requires an understanding of algebra, calculus,

optimization, and other concepts of mathematics. As a result, such models are also known

as algebraic models.

- A classic example of models under this type is the Keynesian national income model for

an open economy, given as:

Y=C+I+G+X–M (1)

C = c0 + c1(Y – T) (2)

T = t0 + t1Y (3)

M = m0 + m1Y (4)

I = I0 (5)

G = G0 (6)

X = X0 (7)

- The solution of the Keynesian national model is obtained by substituting the behavioral

and technological functions into the identity as follows:

Y=C+I+G+X–M

Y = c0 + c1(Y – t0 - t1Y) + I0 + G0 + X0 - m0 - m1Y

Y = c0 + c1Y – c1t0 - c1t1Y + I0 + G0 + X0 - m0 - m1Y

Y - c1Y + c1t1Y + m1Y = c0 – c1t0 + I0 + G0 + X0 - m0

 (1 - c1 + c1t1 + m1)Y = c0 – c1t0 + I0 + G0 + X0 - m0

Equilibrium Y =

- The Keynesian national income model therefore, helps us to estimate the level of

equilibrium national income that an economy should expect, if all the assumptions hold

true.

- The challenge with the Keynesian model is the accurate determination or measurement of

the actual level of national income, and hence it has not been used in measuring income.

Instead, many countries have used the value-added approach in measuring their levels of

national income.

If we assume specific values for the parameters such as: c 0 = 200, c1 = 0.8, t0 = 100, t1 =

0.5, m0 = 60, m1 = 0.2, I0 = 50, G0 = 40, and X0 = 80, then we can obtain the specific value

for equilibrium national income as follows:

Equilibrium Y = = 230/0.8 = 287.5

- Mathematical models are superior to graphical models in two main ways:

(i) They allow us to have more than 2 variables in a model, i.e., they allow for an n-

variable case.

(ii) They allow us to accurately perform sensitivity analysis. Sensitivity analysis is a

method of describing how a system will behave if one or more variables change.

For example, what is the effect of a 10% increase in autonomous consumption on

equilibrium national income?

A 10% increase in c0 = 200 means new c0 becomes 110/100 * 200. Thus, new c0 =

220. We notice that change in c0 = D c0 = 220 – 200 = 20.

Thus, new equilibrium Y becomes:

 Equilibrium Y = = 250/0.8 = 312.5

Thus, change in Y = D Y = 312.5 – 287.5 = 25

Percentage change in Y is thus: %D Y = 25/287.5 * 100 = 8.7 %.

Hence, the effect of a 10% increase in autonomous consumption is an 8.7 percent

change in equilibrium national income.

This exercise would best have been done on EXCEL SPREADSHEET as follows

(Enter the information below into excel spreadsheet):

A B C D E F G H I J K L

1 Scenario c0 c1 t0 t1 m0 m1 I0 G0 X0 Y %D

2 Baseline 200 0.8 100 0.5 60 0.2 50 40 80 K2 -

3 Dc0=10% B3 0.8 100 0.5 60 0.2 50 40 80 K3 L3

4 Dc1=50% 200 C4 100 0.5 60 0.2 50 40 80 K4 L4

5 Dm1= - 200 0.8 100 0.5 60 G5 50 40 80 K5 L5

20%

Assignment 1: perform the following scenario analysis (using both your calculator

and on excel spreadsheet):

 An increase in marginal propensity to consume by 50% on equilibrium

national income. Find D, E and F.

 A decrease in marginal propensity to import by 20% on equilibrium

national income. Find G, H and I.

  If a z% change in government expenditure leads to a 10% increase in

equilibrium national income, what is the value of Z

In cell K2, enter the following formula: = (B2 – C2*D2 + H2 + I2 + J2 – F2) / (1 – C2 +

C2*E2 + G2), then press ENTER. The value reported in that cell should be 287.5

In cell B3, enter the following formula: = 110 / 100 * B2, then press ENTER. The value

reported in that cell should be 220.

To get the value for cell K3, place your cursor in cell K2, then drag-and-drop into cell

K3. The value reported in that cell should be 312.5

In cell L3, enter the following formula: = (K3 – K2) / K2 * 100, then press ENTER. The

value reported in that cell should be 8.7.

Now, proceed that way to get the values in cells: C4, K4, L4, G5, K5 and L5. What

answers do you get in those cells?

(iii) Empirical Models

- Empirical models or econometric models are similar to mathematical models except for

the fact that empirical models heavily rely on data. Therefore, before formulating an

empirical model, one has to gather data on the variables of interest.

- Thereafter, using appropriate statistical techniques, we can then estimate the model from

the data, therefore empirical models calls upon the researcher to have adequate

knowledge of statistics particularly REGRESSION ANALYSIS, e.g., consider the

following data on quantity demanded of a good against versus the price:

 Qd 95 85 75 65 55 45 35 25

P 1 2 3 4 5 6 7 8

Formulate an empirical model from this data. We shall use the regression analysis to

estimate the model: Qd = a + bP

- From regression analysis: the formula for obtaining the Ordinary Least Squares (OLS)

from a simple (bivariate) regression model Y = a + bX + u for the estimate b is: b =

where x = (X – mean of X) and y = (Y – mean of Y), while a is given by a =

ΣY/n - bΣX/n. thus, let Qd be Y and let P be X.

- Enter the data on Y and X on excel spreadsheet column-wise in columns A and B of the

spreadsheet.

- In column C of the spreadsheet type y. Below this cell, which is in cell C2, enter the

following formula: = A2 – 60 and press enter. Now, drag from cell C2 to cell C9.

- In column D of the spreadsheet type x. Below this cell, which is in cell D2, enter the

following formula: = B2 – 4.5 and press enter. Now, drag from cell D2 to cell D9.

- In column E of the spreadsheet type xy. Below this cell, which is in cell E2, enter the

following formula: = C2 * D2 and press enter. Now, drag from cell E2 to cell E9.

- In column F of the spreadsheet type x2. Below this cell, which is in cell F2, enter the

following formula: = D2 * D2 and press enter. Now, drag from cell F2 to cell F9.

- In cell A10, enter the formula: = SUM (A2:A9) and press enter. Now drag across to cell

F10. The values in cells A10 to F10 are the totals or summations.

- In cell A11, enter the formula: = AVERAGE (A2:A9) and press enter. Now drag across

to cell F11. The values in cells A11 to F11 are the means or averages.
- The results are now shown in the table as follows:

Y X Y X xy x2

95 1 35 -3.5 -122.5 12.25

85 2 25 -2.5 -62.5 6.25

75 3 15 -1.5 -22.5 2.25

65 4 5 -0.5 -2.5 0.25

55 5 -5 0.5 -2.5 0.25

45 6 -15 1.5 -22.5 2.25

35 7 -25 2.5 -62.5 6.25

25 8 -35 3.5 -122.5 12.25

ΣY=480 ΣX=36 - - Σxy=-420 Σx2=42

Mean of Y = 480/8 = 60, Mean of X = 36/8 = 4.5

- The table below shows computed values to be used in the regression:

b= and a = ΣY/n - bΣX/n = 60 – (-10 ´ 4.5) = 105

Hence, the regression model: Qd = 105 - 10P

- We can also perform a regression equation using this data on excel spreadsheet. First,

type in the data given column-wise. Go to MENU, click DATA, and then select DATA

ANALYSIS. In the dialogue box that appears, select REGRESSION and click OK. On

input Y range, highlight the data on Y. On input X range, highlight the data on X. Finally

click OK.

- Empirical models can either be linear or non-linear and the specification is usually

informed by economic theory and other econometric tests of model specification such as

the CUSUM test.

 - One advantage of empirical models is that they allow the researcher to perform forecast

or prediction of certain economic variables. For example, we can thus predict what will

happen to quantity demanded when price is 9, and so on. Such predictions are however

valid only if the model is stable and reliable.

(iv) Simulation models

- Simulation means to copy or to mimic a real world phenomenon, so that useful

conclusions can be drawn from the simulation exercise, as what would actually happen in

the real world. This therefore means that, before doing any establishment or development

of an existing system which in most cases are irreversible, simulations help to mimic how

the system would otherwise be and this is less costly and can be easily reversed.

- In a simulation exercise, the model usually begins with an initial equation with values

assigned by the programmer (the baseline scenario), then certain variables are changed –

depending on the issue being examined or the issue of interest or new information

becomes available, and their outcomes or solutions are observed through computer

simulation.

- Thus, simulation models are applicable with graphical, mathematical or empirical models

when the modeler aims to investigate various scenarios.

- Simulation programs therefore can be done on specialized computer software meant for

that exercise.

ADVANTAGES OF SIMULATION:

(i) It is easy to build a model with computer simulation

 (ii) With simulation, we can handle various scenarios and perform scenario analysis, that

involve several assumptions

(iii) For a decision maker, simulation is actually a tool of trade

(iv) Simulation is useful for policy analysis and evaluation

(v) Simulations are useful for comparison purposes

(vi) Changing a real system may be costly and irreversible, but simulation helps to

perform this to mimic how the real system would behave

(vii) With the development of modern computers and specialized software, simulation

exercises have become greatly enhanced, and though expensive in most cases,

cheaper software exists for simulation programming. For instance, excel has features

such as “what-if” and “goal seek” analysis in the toolbar, which are useful for

simulating various scenarios.

DISADVANTAGES OF SIMULATION

(i) Simulation does not guarantee optimality. It is not easy to tell whether the optimal

solution has been reached, or whether it is in fact the best, under the given conditions,

(ii) Consequently, it is difficult to establish the validity of the solution obtained by

simulation exercises,

(iii) The process of simulation requires a lot of expertise to build and maintain, while most

software are also expensive to acquire. However, other cheaper software can be

obtained and training on simulation can be done,

(iv) Simulation does not always yield absolute numbers, and this limits the usefulness of

the output from the simulation programming

APPLICATION OF SIMULATION MODELS

Simulation models have wide applications in day to day life, but they are especially utilized in

the following areas: Manufacturing, health, transport, defense and business process engineering.

These applications are now discussed as follows:

I. MANUFACTURING

- Simulation models are employed in manufacturing system so as to develop and install

manufacturing systems that can deliver high quality products.

- Simulation also allows a company to compare various designs and alternative policies on

the model before actual implementation on the real system.

- By doing this, simulation helps to reduce cost and risk associated with errors. Simulation

has been applied in manufacturing especially in food, beverage and automobiles.

II. HEALTH CARE

- Health care systems aim at providing health care services efficiently and effectively at

minimum cost. To achieve this, health care system test different policies and drugs

without putting the life of patients at risk.

- Hospitals and other health care institutions have applied simulation to determine the best

drug prescription for various ailments; to ensure the movement of patients, staff and

equipment move about in the facility without inconveniencing them or putting life at

danger.

III. TRANSPORT SYSTEMS

 - Transport systems aim to operate as efficiently and effectively as possible and reduce

undue delays to users and goods and services.

- Therefore, simulation models aim at examining various networks or designs and policies

of transport systems so as to continuously improve the services offered by the system.

- Simulation models in transport systems have led to the construction of by-passes and

flyovers, removal of unnecessary roundabouts, expansion and modernization of railway

systems, management of airspace, expansion and modernization of airports and ship

harbors, and so on.

IV. DEFENCE

- Militaries aim at protecting their jurisdictions and repulsing any threats of the enemy as

quickly as possible and using minimum resources.

- They therefore make use of simulation plans to understand various strategies to achieve

these objectives, and counter possible strategies that the enemy could employ.

V. BUSINESS PROCESS ENGINEERING

- In business process engineering, the aim is to improve the performance of a business

entity either to improve service delivery, lower costs of operation, create more effective

management, or improve the customer experience. BPE also aims to ensure that the

business runs its core business efficiently and effectively.

- Examples of Business Process Engineering include: shift from line queuing to electronic

queuing; services in banking halls to agency and internet banking; paper work to

paperless transactions; online tax services (ITAX in Kenya); cash payments to credit card

payments, reduction in the cost of doing business; and so on.

1.2 Evaluating the quality of the model

Many beginning modelers have a tendency to accept the regression estimates as they are

provided by the computer, or those published in journal articles, without focusing on the validity

of those estimates.

However, a modeler should carefully think about and evaluate every aspect of the equations or

model, right from the underlying theory to the quality of data, before accepting the output as

valid. They also have some expectations of the results before the data collection and analysis.

Once the computer estimates have been produced, however, it is then time to evaluate the results.

The following is a checklist of some of the questions that should be asked in evaluating the

quality of the regression estimates:

(1) Is the equation or model specification supported by sound theory i.e., - has the most

theoretically logical functional form been used?

(2) How well does the estimated model fit the data i.e., – to what extent is the model output

consistent with system output?

(3) Is the data set reasonably large and accurate?

(4) Is the estimation technique used (e.g., Ordinary Least Squares - OLS) the best estimator

to be used or other estimation technique(s) would have been appropriate?

(5) How well do the estimated coefficients correspond to the prior expectations by the

researcher?

(6) Are all the important variables included in the model, or are there omitted variables that

may lead to omitted variable bias?

(7) Does the regression appear to be free from major econometric problems that may render

output invalid? Have any model estimator assumptions been violated?

 1.3 Steps in econometric modeling

Although there are no hard and fast rules for conducting econometric research, most

investigators commonly follow a standard method for applied regression analysis:

STEP 1: SELECTION OF THE DEPENDENT VARIABLE

The choice of dependent variable is determined by the purpose of the research, and will be

defined by the topic of the model or research.

STEP 2: REVIEW SCHOLARLY LITERATURE AND DEVELOP THE

THEORETICAL MODEL

Many econometric decisions ranging from which variable to use to which functional form to

employ, are determined by the underlying theoretical model; and these will be informed by

reviewing scholarly work in that area, such as from journal articles.

STEP 3: SPECIFY THE MODEL

Having identified the theoretical model, the modeler may modify the model to derive the

empirical model that will be used. The specification of the model involves choosing the

following components:

- The independent variables and how they should be measured,

- The functional (mathematical) form of the equations, and

- The distribution of the stochastic error term – whether normally distributed or not

normally distributed, such as logistic distribution

Specification of the model should be guided by economic theory.

Any error in the specification is most disastrous to the validity of the estimated model.

STEP 4: HYPOTHESIZE THE EXPECTED SIGNS OF THE COEFFICIENTS

Based on economic theory, we can expect the signs of the coefficients in the model, whether

positive or negative.

We can then compare the regression estimates actual signs against expectations, and this will

have important meaning as far as the quality of the model is concerned.

STEP 5: COLLECT INSPECT AND CLEAN THE DATA

Having specified the model, the analyst then determines the sample size and then collects the

data. If the data is time series, it should be of the same frequency, e.g., monthly, quarterly,

annually, and so on. The amount of data collected should be sufficiently large to ensure enough

degrees of freedom (n – k) will be obtained.

To inspect the data means checking if the data is consistent. For example, a scatter plot of the

data can help to reveal if there are any outliers in the data. Also, summary statistics help us to

examine mean, minimum, maximum, standard deviation, etc all of which will define the data.

Cleaning data involves replacing an incorrect number with a correct one; or where possible, drop

the particular observation from the sample.

STEP 6: ESTIMATE AND EVALUATE THE EQUATION(S)

Choose an appropriate estimation technique, e.g., OLS and software, such as EViews or STATA,

and having estimated the model; evaluate its quality as outlined in the checklist above.

1.4 Uses of Economic Models

Economic models have diverse uses depending on the specific aspects that the modeler is

interested in. in general however; economic models are utilized in the following areas:

i. Models are used to analyze behavior or investigate facts. For example, given a

consumption function, we can then investigate how a change in say tax rate or disposable

income will affect aggregate consumption. Also, how a change in bank rates will affect

borrowing and lending, and so on.

ii. Models can also be used to prove or falsify economic theory. If a theory is tested by

use of a model, and is proven to be false, then such a theory can either be revised in

the light of the available information or it can be replaced by another theory.

iii. Models are useful in evaluating and analyzing policies. Government economic

policies include fiscal policy, monetary policy, trade policy and exchange rate policy.

These policies usually produce various effects, some desirable while other

undesirable. Thus, models can help to examine various policies and to choose among

alternative policies.

iv. Models are very invaluable in measuring and forecasting economic variables, such as

economic growth, and how shocks to the system can affect the variables. This refers

to predicting quantitative values of certain variables typically, the endogenous

variable for the model. For example, the KIPPRA-TREASURY macroeconomic

 model is a long run growth model for Kenya. Apart from growth models, we also

have tax-revenue models, inflation models, exchange rate models, trade models, etc.

Forecasting is meant to guide any policy making unit to achieve its long term goals.

v. Models are also used in structural analysis. This refers to the use of an estimated

econometric model for the quantitative measurement of the underlying

interrelationships of the system under consideration. This involves estimating the

coefficients of the structural equations, reduced form equations, and models with

lagged endogenous variables.

vi. Models are also useful for analyzing strategic behavior among competing economic

agents or policies by regulatory authorities on the industry players. In this regard,

game theory models help to analyze such interactions and access the behavior of

agents affected by other agents or the regulator. For example, how do oil companies

compete for large market share? Or how do oil companies behave following a

regulation by the Energy Regulatory commission on say prices of oil?

vii. With the heavy toll that environmental degradation and climate change take on the

world, most models are trying to incorporate modeling for climate change and its

effects on the economy (green economy models). A famous example of this model is

the T-21 Model developed by the Millennium Institute in the US.

viii. Models have also been constructed for national planning and stabilization purposes.

For example, the World Bank Revised Minimum Standard model (RMSM) and the

IMF’s Polak Model have for long time been used for economic planning in most

African countries. Today however, the IMF’s Financial Programming model has
 taken centre stage in planning and economic stabilization in many developing

countries including Kenya.

Macroeconomic Models and systems, Forecasting, and Policymaking

Models of the macroeconomy have gotten quite sophisticated, thanks to decades of development
and advances in computing power. Such models have also become indispensable tools for
monetary policymakers, useful both for forecasting and comparing different policy options.
Their failure to predict the recent financial crisis does not negate their use, it only points to some
areas that can be improved.

Forecasting and Monetary Policy

Forecasting plays a vital role in the conduct of monetary policy. Policymakers need to predict the
future direction of the economy before they can decide which policy to adopt. While, strictly
speaking, they do not necessarily need an economic model to discuss where the economy is
heading, the use of a model’s forecast has the benefit of elevating that discussion to a scientific
and systematic level. Models can be used to test different theories, for example, and they require
forecasters to clearly spell out their underlying hypotheses.

But policymakers need forecasting tools that do more than project the likely path of important
economic indicators like inflation, output, or unemployment. They need tools that can provide
them with policy guidance tools that help them determine the economic implications of
monetary-policy changes. For example, what will the economy look like under the original
monetary policy, and what will it look like after the change? For this reason, there has been an
effort over the past 40 to 50 years to develop empirical forecasting models that are able to
provide policymakers with this kind of guidance. Three broad categories of macroeconomic
models have arisen during this time, each with its own strengths and weaknesses: structural, non-
structural, and large-scale models.

Structural models are built using the fundamental principles of economic theory, often at the
expense of the model’s ability to predict key macroeconomic variables like GDP, prices, or
employment. In other words, economists who build structural models believe that they learn
more about economic processes from exploring the intricacies of economic theory than from
closely matching incoming data.

Non-structural models are primarily statistical time-series models—that is, they represent
correlations of historical data. They incorporate very little economic structure, and this fact gives
them enough flexibility to capture the force of history in the forecasts they generate. They
intentionally “fudge” theory in an effort to more closely match economic data. The lack of
economic structure makes them less useful in terms of interpreting the forecast, but at the same
time, it makes them valuable in producing unconditional forecasts. That means that they generate
the expected future paths of economic variables without imposing a path on any particular
variable. These unconditional forecasts are typically accurate if the overall monetary policy
regime does not change. Since policy regimes change infrequently, most forecasts from non-
structural models are useful.

The third category, large-scale models, is a kind of middle ground between the structural and
nonstructural models. Such models are a hybrid; they are like nonstructural models in that they
are built from many equations which describe relationships derived from empirical data. They
are like structural models in that they also use economic theory, namely to limit the complexity
of the equations. They are large, and their size brings pros and cons. One advantage is that
relationships can be selected from a huge variety of data series, making it possible to provide a
thorough description of the economic condition of interest. For instance, structural models rarely
feature variables such as “car sales,” while large-scale models often do. The main disadvantage
is their complexity, which poses some limitations to their understanding and use.

System dynamics modelling

System Dynamics Modelling (SDM) is a methodology for studying and managing complex
feedback systems. It is typically used when formal analytical models do not exist, but where
system simulation can be developed by linking a number of feedback mechanisms.

Examples of System dynamics modelling

Examples of its use include modelling the dynamics of the earth's climate, healthcare systems,
the food industry and the military. System dynamics involves causal mapping and the
development of computer simulation to understand system behaviour.

Purpose of system dynamic model

The main goal is to help people make better decisions when confronted with complex, dynamic
systems. The approach provides methods and tools to model and analyzes dynamic systems.
Model results can be used to communicate essential findings to help everyone understand the
system's behavior.

What are the Benefits of Dynamic Modeling?

If properly performed, Dynamic Modelling can reveal design flaws that may not show up readily
during the prototyping and testing phases of the product design cycle.

Unique benefits that dynamic modelling provides include:

 Identifying interactions between subsystems of a complex product which may be too

expensive to create during physical prototyping and testing,
 Identifying potential failure modes which should be tested in physical prototypes, before
hard tooling,
 Simulating dynamic loadings which may be difficult to create during actual testing,
 Identifying functional limitations on the use of a product.

Econodynamics

Econodynamics is an empirical science that studies emergences, motion and disappearance

of value a specific concept that is used for description of the processes of creation and
distribution of wealth.

Dynamism is a general name for a group of philosophical views concerning the nature of
matter. However different they may be in other respects, all these views agree in making matter
consist essentially of simple and indivisible units, substances, or forces.
Dynamic economics has an important place in economics because many economic theories are
based on it. For example, saving and investment theory, theory of interest, effect of time
element in price determination, etc. are based on dynamic economics.

The competition and job-to-job mobility fostered by dynamism helps power wage gains for
workers. In a churning labor market where demand for workers is high and competition is
intense, workers gain leverage and have ample opportunities to seek out better jobs and higher
wages.

Each economy functions based on a unique set of conditions and assumptions. Economic
systems can be categorized into four main types: traditional economies, command economies,
mixed economies, and market economies

Decision theory

Decision theory deals with methods for determining the optimal course of action when a number

of alternatives are available and their consequences cannot be forecast with certainty. It is

difficult to imagine a situation which does not involve such decision problems, but we shall

restrict ourselves primarily to problems occurring in business, with consequences that can be

described in dollars of profit or revenue, cost or loss. For these problems, it may be reasonable to

consider as the best alternative that which results in the highest profit or revenue, or lowest cost

or loss, on the average, in the long run. This criterion of optimality is not without shortcomings,

but it should serve as a useful guide to action in repetitive situations where the consequences are

not critical. (Another criterion of optimality, is the maximization of expected utility, provides a

more personal and subjective guide to action for a consistent decision-maker.)

The simplest decision problems can be resolved by listing the possible monetary consequences

and the associated probabilities for each alternative, calculating the expected monetary values of

all alternatives, and selecting the alternative with the highest expected monetary value. The

determine of the optimal alternative becomes a little more complicated when the alternatives

involve sequences of decisions.

In another class of problems, it is possible to acquire often at a certain cost additional

information about an uncertain variable. This additional information is rarely entirely accurate.

Its value hence, also the maximum

amount one would be willing to pay to acquire it should depend on the difference between the

best one expects to do with the help of this information and the best one expects to do without it.

Decision modelling stages

1. Identify Decisions.

 Identify the decisions that are the focus of the Next-Best-Action platform.
o From a marketing perspective these decisions usually fall into three classes:
eligibility, relevancy and prioritisation
 What are the KPIs (Key Performance Indicators) surrounding that decision? Now is the
time to consider how a decision will drive a given KPI such as churn, risk, NPS, etc.

2. Describe Decisions.

 Describe the decisions and document how improving them will impact the business
objectives and metrics of the business. The activities performed in this stage would guide
how simulations are performed on the strategy. For example, what would be the likely
impact of using a suitability rule which excludes propositions related to already owned
 products? What would be the impact of using a different prioritisation algorithm which
considers NPS (Net Promotor Score) for service-related communications?

3. Specify Decision Requirements.

 Specify the detailed information and knowledge required to make the decisions. What are
the inputs (human and computer) required to make the decision? Decision requirements
will now be used to express this. For example, what knowledge would be required to
define suitability rules for a cross-sell mobile international call discount proposition? The
cross-sell marketing leader would provide input to determine who should (or should not)
be suitable for the international call discount proposition.

4. Decompose and Refine the Model.

 Refine the requirements for these decisions using the graphical notation of Decision
Requirements Diagrams. In addition, identify additional decisions that need to be
described and specified.
 For example, eligibility decisions could be decomposed by applicability and suitability.
Applicability defines rules which ask “Can I show this proposition?” Suitability defines
the rules which ask “Should I show this proposition?”
 The cycle is repeated. We repeat the process by returning to step 1.

Macroeconomics in a Self-Organizing Economy

Our starting point is one of the oldest and most important ideas in economics, going back at least
to Adam Smith, namely the idea that a decentralized economic system is self-organizing. It is
capable of “spontaneous order,” in the sense that a globally coherent pattern of transactions can
result from purely local interactions, without the intervention of a central coordinator. Indeed,
like an anthill, a free market economy can organize transactions into patterns that are beyond the
comprehension of any of its individual participants. We would like to understand how this self-
organization takes place. Specifically, what is the process that coordinates the exchange activities
of millions of independent transactors in a decentralized economy?
The reason why these questions are critical for understanding macroeconomic policy is that an
economy’s coordination mechanism works better at some times than others. Even Smith and
Hayek recognized that the automatic workings of the decentralized economy could sometimes be
improved by collective intervention. Consider, for example, the increase in unemployment that
takes place during a deep recession. Unemployed workers who used to be employed are just as
willing and able to work as before, the fall in aggregate output that accompanies recession has
enhanced the scarcity value of the output they could potentially produce if employed, and yet the
market for their services has somehow disappeared. The coordination mechanism that had
previously allowed them and those with a taste for their output to realize their potential gains
from mutually advantageous exchange is no longer allowing them to do this, even though those
gains are if anything larger than before. Macroeconomic policy to deal with unemployment thus
amounts to fixing a mechanism that has malfunctioned, and a highly complex mechanism at that.
And attempts to fix this broken mechanism without first understanding how it is supposed to
work normally are likely to be as successful as medieval medicine was in treating bacterial
infections.

The main premise of research has been that the role of coordinating transactions in a
decentralized economy is performed, for better or worse, by a self-organizing network of
business firms that seek profit from creating and operating the markets through which others
transact. To use a phrase that Clower once coined, business firms are the visible fingers of the
invisible hand. Economics has a long tradition of regarding exchange as a do-it-yourself affair, in
which people with goods and services to sell trade directly with the ultimate demanders of those
goods and services. But a little reflection on the experience of daily life is enough to persuade
most people that exchange in a market economy is not a do-it-yourself affair. People are not like
the actors in a typical monetary search model, who when hungry go wandering aimlessly in
hopes of randomly encountering someone with surplus food. They go to a grocery store or to a
restaurant. When in search of clothing they visit a tailor or a clothing store. They lend surplus
funds through the intermediation of a bank, arrange for long-distance travel by using facilities
provided by a travel agent, and so on and so forth. Most of us also sell our labor services to an
economic entity, either a private business or a government agency (the latter of which would not
exist in a purely decentralized economy) whose primary purpose is to purchase various such
services, organize them into production, and sell the resulting output. So to understand how, and
how well, exchange activities are coordinated in a decentralized economy, the first place to look
is to this self-organizing network of firms that constitutes the institutional structure through
which we all conduct our daily business.

Now some would argue that a good economic theory involves abstraction, and that if we want to
model the transactions process of a modern economy we should maybe abstract from business
firms, by assuming that people who work for businesses trade their output directly with those
having a taste for the workers’ output. This is the stance taken by search theoretic models of
money, which typically assume that trade takes place directly between ultimate suppliers and
ultimate demanders, who meet in random non-repeated encounters without the aid of any
intermediary. Presumably the rationale for this way of looking at the transactions process is the
same as the rationale for abstracting from money in the theory of value. On the surface, what we
see is people trading goods and services for money, but the deeper underlying reality that we see
once we pierce the veil of money is that people are ultimately trading goods and services for
other goods and services, with money acting only as a device for executing these ultimate
exchanges.

The analogy between money and firms is a useful one. But as John Stuart Mill once observed,
there is nothing more insignificant than money, except when it goes wrong. By the same token,
the fact that people find it convenient to trade through shops rather than directly with one another
is perhaps of little significance for understanding the long-run structure of relative prices. But
when something goes wrong with the network of firms that people normally rely upon, then
abstracting from the existence of such firms is as unhelpful as it would be to ignore money when
trying to understand inflation.

Moreover, a good case can be made that, by recognizing that production takes place in firms but
not recognizing their role in coordinating transactions by creating and operating markets, we are
ignoring their most important activity, as measured by the value of resources they use. The value
added in Finance, Insurance, and Real Estate, for example, is typically much more than that of
the entire manufacturing sector, as is the value added in Retail and Wholesale Trade. Moreover,
much of the input to the manufacturing sector is best construed as being used up in the
production of transactions that help people realize gains from trade rather than being used up in
transforming inanimate objects. We have in mind the inputs of lawyers, sales people, those
engaged in personnel, marketing, and advertising, and so forth, all of whom are undertaking
activities whose main purpose is to facilitate transactions.

We were referring, of course, to the broad class of rational-expectations equilibrium models

generally known as DSGE, for dynamic stochastic general equilibrium. DSGE started out 4
decades ago as a reaction against the Keynesian economics that had been the dominant paradigm
of the profession in the 1950s and 60s, a reaction that was first expressed by new classical
economists, like Lucas and Sargent, and later by real business cycle theorists. The early
proponents of DSGE argued that an equilibrium model built on a slightly modified Walrasian
conceptual framework, in which markets clear everywhere and always, could account for
important short-run as well as long-run macroeconomic phenomena. But soon Keynesian
economists began developing their own versions of DSGE, which consisted of rational
expectations equilibrium models in which not all prices were perfectly flexible, and by now
DSGE has become the dominant paradigm agreed upon by all sides of the great macroeconomic
debates.

Of course, there are many serious criticisms one might make of DSGE, and many of them have
been made in the literature. The criticism we consider most important for present purposes is that
existing DSGE models, even those with imperfectly flexible prices, are built on a conceptual
foundation that pays little or no attention to the way in which economic transactions are
organized. To borrow a phrase from Jevons, they ignore the institutions and processes that make
up the mechanism of exchange.

When we examine DSGE models, looking for what else might go wrong with the market
mechanisms that coordinate economic transactions, we find that in most of them there is no such
mechanism. In models with perfect competition, the setting of prices is left to a mysterious
outside agent called the auctioneer, whose behavior is left largely unexplained. But that is just
the tip of the iceberg. There is no description of how trades are arranged. Even if we accept that
the auctioneer can provide everyone with a price vector such that the sum of desired demands
equals total supply for each tradable object, there is no account of how buyers and sellers are
matched up with one another and how the trades that people have planned will be executed.
When demand and supply are not equal, the theory offers no guidance as to who gets to trade
how much and with whom, no indication of how people learn about trading opportunities, about
who creates and maintains the shops and other facilities through which they trade, about how
bids and offers are transmitted, and so on and so forth.

The canonical model of Woodford (2003), which forms the basis of the estimated New
Keynesian DSGE models, now used in central banks around the world, makes less use of the
mysterious auctioneer, inasmuch as many prices are set by a given set of monopolistically
competitive firms who are explicitly motivated to maximize their shareholders’ wealth. So far, so
good. But, there is no account of where these price-setting monopolists come from, how they
maintain their monopolies against the threat of entry, how people decide to trade with one set of
firms rather than another, how firms manage to coordinate with their suppliers and customers,
what happens when one of them goes out of business in a recession, and so forth. Instead, all
transactors are in continuous touch with each other through the intermediation of these firms,
whose existence is merely assumed, and who take care of enough details of the transactions
process that the other people in the model are connected only through the market prices that they
take as given from the firms. As a result, there is nothing that can go wrong in the transactions
process other than some mistake in price-setting.

In essence, these New Keynesian DSGE models are providing the same diagnosis that
economists have given for generations; unemployment rises because wages and prices are slow
to adjust to shifts in demand and supply. This is the answer provided by classical economists
from Hume through Marshall. It is still the answer offered by modern Keynesian economics.
Indeed, it is now even the answer that has been finally accepted by most proponents of the real-
business-cycle school of macroeconomics, who admit the need for wage-and-price stickiness to
account for various features of the business cycle.

The problem with this time-honored tradition of blaming wage-and-price stickiness is not that
the assumption of stickiness is factually incorrect. On the contrary, the stickiness of wages and
prices is one of the most well-documented facts of macroeconomics. Instead, as Leijonhufvud
(1968) forcefully pointed out, the problem is that, first, the experience of the Great Depression in
the United States shows clearly that the downturn that started in 1929 did not come to an end
until wages and prices started to rise, that is, until the reflation that was clearly a deliberate
policy move on the part of the Roosevelt administration started to take place. If lack of wage and
price flexibility had caused the downturn, then it would have taken deflation rather than reflation
to cure the unemployment problem. Second, as Keynes argued in Chapter 19 of the General
Theory, and as Fisher had already argued in Debt Deflation Theory of Depressions, there are
many reasons for believing that wage and price flexibility would actually make fluctuations in
unemployment larger rather than smaller.

So when unemployment rises in a recession, something has gone wrong with the process by
which economic transactions become organized, something that goes beyond the mere stickiness
of wages and prices, something that we think can only be discovered by investigating simple
stylized models of economies in which trading activities take place, in and out of equilibrium, in
a self-organizing network of markets that are created and operated by profit-seeking business
firms, and by asking how, and how well, those activities are coordinated in various
circumstances. What we would like to do in this paper is to give the reader an idea of what kind
of model that research has led us to construct, and why we think this class of models provides a
more solid framework for analyzing certain policy questions than does any DSGE model
currently in use.

It turns out that this research agenda is one for which Agent-Based Computational Economics
(ACE) is particularly well suited for two main reasons. First, by endowing each agent with a set
of relatively simple adaptive behavioral rules that allow the agent to operate intelligently in an
unknown environment, an ACE model gives the system a chance to achieve some semblance of
order without giving anyone the kind of systemic knowledge that would allow him to act as a
central coordinator. A rational expectations equilibrium might or might not emerge from the
interaction of these rules. If it does, then we have discovered something about at least one
possible mechanism that produces that kind of spontaneous order whereas, if the system fails to
approximate a rational expectations equilibrium, we will have discovered something about the
conditions under which a spontaneous order is likely to require some kind of collective
intervention. The second reason for using ACE is that models of spillovers between multiple
markets that are not in supply-demand equilibrium are notoriously difficult to analyze. The
attempts by Barro and Grossman, Benassy, Malinvaud, and others in the 1970s to understand
what some called “general disequilibrium analysis,” building on the original contributions of
Clower and Patinkin, made little progress largely because the problem quickly became
analytically intractable. ACE can deal with this kind of intractability by substituting simulation
and Monte-Carlo results for unattainable analytical results.

1.5 ACTIVITY ONE

Read and make notes on the following models:

(i) The World Bank Revised Minimum Standard Model (RMSM)

(ii) The IMF Polak Mode

(iii) The IMF Financial Programming model

(iv) The Threshold-21 (T-21) Model by the Millennium Institute

(v) The KIPPRA-Treasury Macroeconomic model

For each model, follow the following outline:

(a) What the model is all about and uses of the model

(b) The assumptions behind the model

(c) The structure of the model – i.e., variables, equations, etc

(d) Application of the model (where has it been applied)

(e) Limitations of the model

References/Reading List
1. Neradilová, H., & Fedorko, G. (2016). The use of computer simulation methods to reach
data for economic analysis of automated logistic systems. Open Engineering.
2. Brown, C. (2021). Quantitative modelling and computer simulation. In The Routledge
Handbook of Landscape Ecology. Routledge.
3. Bossel, H. (2018). Modeling and simulation. AK Peters/CRC Press.
Supplementary Reading List

1. Albright, Christian, Christopher J. Zappe and Wayne L. Winston (2011). Data Analysis,
Optimization and Simulation Modelling: Australia, Cengage learning
2. Rajib Mall (2010). Fundamentals of software engineering: New Delhi, PHI learning
3. Aluja, J. G. (2012). Towards an advanced modelling of complex economic phenomena.
Berlin, Heidelberg: Springer Berlin Heidelberg.
4. Taudes, A. (2005). Adaptive Information Systems and Modelling in Economics and
Management Science. Vienna: Springer Vienna