INTRODUCTION TO ECONOMETRICS

What is Econometrics?

The term ‘Econometrics’ was first coined by Norwegian economist and statistician Ranger Frisch in 1926.
Econometrics is the science which integrates three subjects – economic theorems, statistics and mathematical
science for the purpose of investigating the empirical support of economic theorems. Though it is the amalgamation
of three subjects (Economics, Statistics and Mathematics), but it is completely different from each one of them.

Why Econometrics to be taught as a separate subject?

i. In many cases economic theory postulates the relationship between two or more variables not in exact
quantitative terms, rather slightly quasi-quantitative term. For example, the demand function for a product states
that the demand for the product is a function of price of the product itself, prices of other related products
(substitutes and/or complementary), income of the consumers, taste and preferences of the consumers and many
other related factors. More specifically, it can be said that with all other factors remaining the same, the quantity
demanded is inverse related with the price of the product. However, the economic theorem doesn’t tell about the
strength of such inverse relationship. That it, it ignores the exact specification of the demand function. Similarly, the
investment multiplier is macro-economic theorems tell that that the gross domestic product will increase following
an increment in investment expenditure through the multiplier mechanism - the more the value of marginal
propensity to consumer, the more the value of multiplier effects. Neither economic theorems nor the mathematical
model nor statistics measures this effects empirically. Pure economic theorems do not tell us much about the
amount of changes in the target variable (demand for product or gross domestic product in this case) due to some
changes in its underlying variables. Here econometrics comes as handy as it provides such information.

ii. Mathematical economics explains the economic theorems in terms of mathematical forms (equations) without
much emphasis or no emphasis on its measurability and empirical verification. However, econometrics seeks to find
the empirical evidences and verification of mathematical forms of economic relationship. So, it requires empirical
skill as well as theoretical knowledge of the subject.

iii. Statistics in economics does the same purposes it the subject is meant for, that is, collection, compilation,
processing and presentation of the economic data. This kind of the data (for example, Gross Domestic Product,
inflation rate, unemployment rate, balance of payments etc.) so accumulated will not be able to speak much about
the true facts of the economy until and unless the data is dealt with econometrics tools. In other words, economics
data gets life into it when it is treated with econometric tools.
Methodology of Econometrics

The sequence of econometrics methodology can best be understood with the help of the following flow chart:

Framing the
Hypothesis

Mathematical Model

Econometric/statistical
model

Data
collection/compilation

estimatin of the model

checking the
hypothesis

validatin of the model

policy prescription

Desirable Properties of an Econometric Model

1. Theoretical Plausibility

The model should be compatible with the postulates of economic theory. It must describe adequately the economic
phenomenon to which it relates. There should not be any lag in describing all the detailed about economic theorems
with the estimated model.

2. Explanatory Ability

The model should be able to explain the observations of the actual real world. It must be consistent with the
observed behaviour of the economic variables whose relationship it determines.

3. Accuracy of the Estimates of the Parameters

The estimates of the coefficients should be accurate in the sense that they should approximate as best as possible
the true parameters of the structural model.

4. Forecasting Ability

The econometric model should produce satisfactory prediction of the future values of the dependent variables based
on the values of the independent or explanatory variables.

5. Simplicity

The econometric model should represent the economic relationship among the variables with maximum simplicity.
The model with too many complexity and vagueness will create problem in the estimation of the parameters and
hence on prediction.

Use of Computer Software

The entire aspects of econometric application would be possible if the amalgamation of these components is
executed through the computer software programming. In fact, the application of modern computer software
programme has been become an integral part of applied econometrics. There are numerous computer software are
available which can be classified into two categories – proprietary and open-source software. In the closed source or
proprietary categories, Stata, SPSS, Limdep, Shazam, Eviews, SAS, Microfit, Minitab are the most popular. On the
other hand, R and Gretl are the two most popular software in the open-source category.

In this book, detailed discussion about the R programming will also be made along with the theoretical foundation of
the subject of econometrics.

Measurement Scale of Variables

Most of the economic variables will fall into four broad categories: ratio scale, interval scale, ordinal scale and
nominal scale. Variables are classified on the basis of three criteria – ratio, distance and ordering. Brief description of
each of these types are explained as under:

Ratio Scale

A variable X is said to be ratio scale variable if it satisfies all the three criteria above. More specifically, for any two
values X1 and X2 of the variable X, the ratio X1/X2 and the distance (X1 − X2) are meaningful quantities. Also, the
natural ordering (ascending or descending) of the values of the variables is possible, that is, either X1 < X2 or X2 < X1
is meaningful. Most of the economic variables fall into this category. For example, GDP, Investment, Consumption,
Saving etc. all are ratio scale variable.
Interval Scale

A variable which satisfies the last two properties of ratio scale variable but not the first one, is known as Interval
Scale variable. That is, ratio of any two values of the variable is meaningless. For instance, time (year, month, date
etc.) is a very good example of this category. You can find the difference two time points such as the difference
between 1991 and 2018 is 17 years. Also, you can order the time period in ascending order or descending order as
per your convenience. However, it will not be meaningful to take the ratio of years point. That is, the ratio 1991 2018
is meaningless. However, the difference (2018 − 1991) and their ordering in terms of occurrence 1991 > 2018 is
meaningful.

Ordinal Scale

A variable is said to be ordinal scale if it satisfies only the third property of the ratio scale (i.e., natural ordering). For
examples, the grading systems (A, B, C grades) or income class (upper, middle, lower) are the best ways to describe
certain economic features of population. For these variables the ordering exists but the distances between the
categories cannot be quantified as also the ratios.

Nominal Scale

Variables in this category have none of the features of the ratio scale variables, rather the variables do have some
inbuilt characteristics for grouping or classifying them. Variables such as gender (male, female) and marital status
(married, unmarried, divorced, separated) simply denote categories. Please keep in mind that econometric method
will vary depending on the nature of the variables involved in the study. Econometric methods which have been
designed for ratio scale variable, may not be very suitable for categorical or interval scale variable. For this reason
you must be very careful about the nature and structure of the data being used in the analysis.