Econometrics
An Introduction
Literally interpreted, econometrics means “economic measurements”. Although measurement is an
important part of econometrics, the scope of this discipline is much broader.
Why a separate discipline?
Econometrics is an amalgam of economic theory, mathematical economics, economic statistics and
mathematical statistics. Yet the subject deserves to be studied in its own right because of the
following reasons:
1. Economic theory makes statements or hypotheses that are mostly qualitative in nature. For
example, economic theory states that, other factors remaining unchanged, a reduction in price,
in general, leads to an increase in the demand of a commodity. But the theory itself does not
specify any mathematical relationship between the two, i.e., it does not specify by how much
the quantity demanded will increase for one unit fall in price. It is the job of the
econometrician to provide such numerical estimates.
2. The main concern of mathematical economics is to express economic theory in mathematical
form (equations). However, it does not do any empirical verification of these mathematical
models. Econometrics is concerned with empirical verification of these models through data.
Note: In econometrics, we are mainly concerned with observational data, rather than, empirical
data.
Methodology of Econometrics:
Broadly speaking, traditional econometric methodology proceeds along the following line:
Statement of theory or hypothesis.
Specification of a mathematical model for the theory.
Specification of the statistical or econometric model.
Obtaining the data.
Estimation of the parameters of the model.
Hypothesis testing.
Forecasting or prediction.
Statement of theory or hypothesis
An economic hypothesis may be – “Consumption expenditure is directly related to disposable
income”.
Specification of the mathematical model
The above hypothesis does not specify any precise mathematical form between the variables
consumption expenditure (y) and disposable income (x). For simplicity, a mathematical economist
might suggest a relation of the form
y=β 1 + β 2 x … ..(1)
Specification of the model
Equation (1) assumes that there is an exact or deterministic relationship between x and y. Thus,
according to (1), if there are 20 persons having the same disposable income, they will show exactly
same consumption pattern. However, in reality, the case is not like this, because, in addition to
income, other factors like family size, age of family members, occupation, etc. are likely to affect
expenditure. To allow for the inexact relationship, the econometrician would modify (1) as below:
y=β 1 + β 2 x +u … ..(2),
where u, known as the disturbance term or error term or white noise, is a random variable that has
well defined probabilistic properties. (2) is an example of econometric model.
Obtaining data
Data are necessary to estimate the parameters involved in the model.
Estimation of the parameters
On the basis of the collected data, one can now estimate the parameters β 1 and β2 involved in the
model using some appropriate method of estimation. Let ^
β 1 and ^
β 2 be these estimates. The estimated
model is then obtained as
Y=^
β1 + ^
β 2 x … .(3)
where Y is the predicted or regressed value of y.
Hypothesis Testing
The next step is to test whether the proposed model is in agreement with the underlying theory. We
thus need to perform tests of hypotheses concerning the parameters involved in the model.
Forecasting
Once the proposed model is found to be satisfactory, it may be used to predict the value of y for given
values of x.