CHAPTER 1
Introduction Econometrics
What is econometrics?
• Literally speaking, the word ‘econometrics’ means measurement in economics.
• In general, econometrics is the application of statistical and mathematical methods to the analysis of
economic data with a purpose of giving empirical content to economic theories and verifying or
refuting them.
• More specifically, it is concerned with the use of statistical methods to attach numerical values to
the parameters of economic models and also with the use of these models for prediction.
• The techniques of econometrics consist of a blend of economic theory, mathematical modelling and
statistical analysis.
Before any statistical analysis with economic data is performed, one needs a clear mathematical
formulation of the relevant economic theory. For example, saying that the demand curve is downward
sloping is not enough. We have to write the statement in mathematical form as:
q = α + βp ,β < 0 or
q = α pβ ,β < 0
where q is the quantity demanded and p is the price.
One major problem: economic theory is rarely informative about functional forms. Thus, we have to
use statistical methods to choose the functional form as well. Putting it differently, we are often
presented with no more than the data themselves and the theory behind the data generation process is
non-existent or far from being complete. Thus, what we do in practice is:
• Investigate the important features of the observed data
• Construct an empirical model (incorporating as much available background theory as possible)
• Check that the constructed model is capable of capturing these important features
The model construction phase is facilitated by specifying a fairly wide class of models within which
some optimal search technique may then be applied.
The main aim of this course is to provide a good understanding of the properties of linear econometric
models and techniques. Although some features of macroeconomic time series can not be adequately
described and analyzed using linear techniques, much econometric model building is dominated by
linear models. The reasons include:
• The theory of statistical inference is the most developed for linear models.
• Linear approximations to economic relationships have been quite successful in empirical work.
Economic and econometric models
The first task an econometrician faces is that of formulating an econometric model.
What is a model?
A model is a simplified representation of a real-world process. For instance, ‘the demand for oranges
depends on the price of oranges’ is a simplified representation since there are a host of other variables
that one can think of that determine the demand for oranges. These include:
• Income of consumers
• An increase in diet consciousness (e.g. drinking coffee causes cancer; so better switch to orange
juice)
• Increase or decrease in the price of substitutes (e.g. that of apple)
However, there is no end to this stream of other variables! Many have argued in favour of simplicity
since simple models are easier:
• To understand
• To communicate
• To test empirically with data
The choice of a simple model to explain complex real-world phenomena leads to two criticisms:
• The model is oversimplified
• The assumptions are unrealistic
For instance, to say that the demand for oranges depends on only the price of oranges is both an
oversimplification and an unrealistic assumption.
• To the criticism of oversimplification, many have argued that it is better to start with a
simplified model and progressively construct more complicated models.
• As to the criticism of unrealistic assumptions, the relevant question is whether they are
sufficiently good approximations for the purpose at hand or not.
In practice we include in our model:
• Variables that we think are relevant for our purpose.
• A ‘disturbance’ or ‘error’ term which accounts for variables that are omitted as well as all
unforeseen forces.
This brings us to the distinction between an economic model and econometric model.
An economic model is a set of assumptions that approximately describes the behaviour of an economy
(or a sector of an economy).
An econometric model consists of the following:
a) A set of behavioural equations derived from the economic model. These equations involve some
observed variables and some ‘disturbances’.
b) A statement of whether there are errors of observation in the observed variables.
c) A specification of the probability distribution of the ‘disturbances ‘.
With these specifications, we can proceed to test the empirical validity of the economic model and use it
to make forecasts or use it in policy analysis.
Econometrics Lecture notes for 2
Masters Program Hawassa University
Methodology of econometric research
The aims of econometrics are:
a) Formulation of econometric models, that is, formulation of economic models in an empirically
testable form (specification aspect).
b) Estimation and testing of these models with observed data (inference aspect).
c) Use of these models for prediction and policy purpose.
Schematic description of the steps involved in econometric analysis (research)
Economic theory
Econometric model Data
Estimation
Specification testing and
diagnostic checking
Is the model adequate?
NO YES
Test of any hypotheses
Using the model for
predictions and policy
Econometrics Lecture notes for 3
Masters Program Hawassa University
Description of the steps
1. We begin with an economic model which is a set of assumptions that describes the behaviour of an
economic phenomenon.
2. Formulation (specification) of an econometric model: a set of equations derived from the
economic model that involve some observed variables and some ‘disturbances’.
3. Collection of relevant data on variables implied by the econometric model.
4. Estimation of model parameters using mathematical statistics and probability theory.
5. We conduct tests to verify whether:
• The specification of the model is correct
• Model assumptions are valid
6. Based on step (5):
• If the model failed to pass the specification testing and diagnostic checking step, then one has to
revise the specification of the econometric model (or new specification).
• If the model passes the specification testing and diagnostic checking step, then one has to
proceed with testing any hypothesis of interest (e.g. which of the explanatory variables
significantly affect the response (endogenous) variable?).
7. We use the estimated model for predictions and policy.
Note: It is not true that econometricians just take the theories they are given and test them, learning
nothing form the tests. Based on the results, they either verify or refute the economic theory or suggest
alternative theories.
Econometrics Lecture notes for 4
Masters Program Hawassa University
