Unit 1

Introduction: Nature, Scope &

Methodology of Econometrics.
TABLE OF CONTENTS

SL. Topic Fig.N SAQ/ Pag

No. o. Activi e
/Table ty No.
/
Graph
1.0 Introduction
1.1 Learning objectives
1.2 Nature and meaning of
Econometrics
Figure 1.1
1.2 Characteristics of econometrics
.1
1.2 Relationship between Economic
.2 theory, Mathematical economics,
and econometrics
1.2 Types of econometric analysis
.3
Self-Assessment Questions 1
1.2 Structure of economic data
.4
1.2 Types of Economic questions
.5
1.2 Framework for approaching
.6 econometric analysis
Self-Assessment Questions 2
1.3 Scope of econometrics
1.3 Examples of Research problems
.1
1.3 Application of econometrics in
.2 specific fields
Self-Assessment Questions 3
1.4 Methodology of econometric
research
1.4 Assumptions in Econometric
.1 modelling
Figure 1.2
1|Page
1.4 Steps in econometric analysis
.2
Self-Assessment Questions 4
1.5 Limitations of Econometrics
1.6 Future Relevance and Emerging
Areas in Econometrics
Self-Assessment Questions 5
1.7 Conclusions
1.8 Summary
1.9 Glossary
1.1 Terminal Questions
0
1.1 Key answers
1
1.1 References and Further reading
2
1.0 INTRODUCTION
Economic research, particularly in fields like economics, finance, management,
and marketing, is becoming increasingly quantitative. As a result, students
entering these disciplines are often encouraged—or even required—to study
econometrics. As the name suggests, Econometrics aims to measure economic
relationships by making the use of economic theory, mathematical economics,
economic statistics (i.e., economic data), and mathematical statistics.

The term "econometrics" was first introduced in 1926 by Ragnar A. K.

Frisch, a distinguished Norwegian economist. Alongside Jan Tinbergen, another
pioneer in the field, Frisch was awarded the first Nobel Prize in Economic
Sciences in 1969. Although many economists had been working with data long
before the introduction of the term, Frisch emphasized the need for a distinct
discipline that systematically applies statistical techniques to economic analysis.
Today econometrics is a vast field within economics that continuously evolves
with the introduction of new tools and techniques. It focuses on quantifying
economic relationships by integrating economics, mathematical economics, and
statistics to assign numerical values to the parameters of economic
relationships. Economic theories are typically represented in mathematical
forms and combined with empirical data. Econometric methods are employed to
determine parameter values, which correspond to the coefficients in the
mathematical representation of economic relationships. These relationships
capture the random behaviour of economic interactions, an aspect generally
overlooked in standard economic and mathematical models. Econometric
modelling is a crucial aspect of econometrics that involves the formulation and
2|Page
estimation of mathematical models to analyse economic relationships. These
models use real-world data to test economic theories, forecast future trends, and
evaluate policy impacts. A typical econometric model consists of dependent and
independent variables, where statistical techniques such as regression analysis
help determine the strength and direction of relationships between them. The
reliability of an econometric model depends on the accuracy of data, the validity
of underlying assumptions, and the appropriateness of the chosen methodology.
By transforming theoretical concepts into measurable equations, econometric
modelling provides a systematic approach to empirical economic analysis,
making it an essential tool for researchers, policymakers, and financial analysts.

This chapter provides an in-depth exploration of econometrics, covering

its definition, scope, methodologies, and applications. It distinguishes
econometrics from mathematical economics, highlighting their
interconnectedness yet distinct roles in economic analysis. Additionally, the
chapter discusses the practical use of econometrics in various fields of
economics and examines the challenges associated with its implementation. By
understanding the methodologies and limitations of econometrics, we gain
insight into its significance in both theoretical and applied economic research.

1.1 LEARNING OBJECTIVES

By the end of this chapter, you will be able to:

 Understand fundamentals of econometrics and differentiate it from

mathematics and statistics.
 Identify the nature and key characteristics of econometric analysis.
 Examine various scope and applications of the subject.
 Analyse methodologies used in econometrics.

1.2 NATURE AND MEANING OF ECONOMETRICS

What is econometrics?
Econometrics is the science and art of using economic data to measure the
empirical validity of economic theory by applying mathematical and statistical
tools. Econometrics is considered a scientific analysis because it involves
observing specific phenomena, collecting relevant data, analysing that data, and
ultimately drawing conclusions. Econometrics deals with the measurement of
economic relationships. Econometricians usually works with non-experimental
data in which variables are not controlled like in laboratory experiments. The
significance of applied work in economics is increasing constantly.
3|Page
 Economics Study economic data

Econometrics= Econ+ USES TO

Metrics Mathematics Test hypothesis

Statistics Make future predictions

Fig:1.1 illustrates the basic framework of econometrics and how it uses different disciplines
to study economic data, to test hypothesis, and to make future predictions.

Econometrics provide enough empirical information to economic theories and

helps to study economic data, to test hypothesis, and to make future
[Link] transforms the qualitative economic content into measurable or
quantifiable data for making economic analysis more scientific and precise.
1.2.1 Nature and characteristics of economics
1. Interdisciplinary Nature: Integrates the fields of economics,
mathematics, and statistics to study the economic phenomena.
2. Empirical Analysis: Emphasizes an empirical and data-driven approach
that relies on data from the real world to test ideas, hypothesis, and
models.
3. Data from non-Experimental Contexts: Constructs model using data
collected in an uncontrolled, realistic context, as opposed to experimental
or controlled contexts.
4. Model Building: It builds mathematical and statistical models to describe
and express economic relationships.
5. Hypothesis Testing: It tests economic generalizations and models using
statistical tests.
6. Quantification of economic relationships: As the name suggests it
measure the strength and direction of relationships between two or more
economic variables
7. Ability to forecast as well as policy evaluation: helps in forecasting future
trends using past data and assists governments as well as organisations in
evaluating and designing effective economic policies.
1.2.2 Relationship between Economic theory, Mathematical economics, and
econometrics
Economic theories are often shown in mathematical forms and paired
with real-world data. These theories are the frameworks that explain how
4|Page
economies function and how different factors interact within an economic
system. Economic theories provide insights into production, consumption,
distribution of wealth, and market behaviours. They help economists and
policymakers understand economic trends, predict future developments, and
design strategies to address economic issues. Economic theories form the
foundation for econometrics and mathematical economics, which help in
analysing and testing these theories using quantitative methods. The term
Mathematical Economics can sometimes be mixed up with the term,
Econometrics. As the ‘metric’ part of Econometrics suggests, it mainly focuses
on measuring economic data. This means it looks at the study of real-world
observations using statistical methods for estimation and hypothesis testing.
Mathematical Economics, on the other hand, is about using mathematical tools
for interpreting theoretical side of economic analysis, with little or no focus on
issues like measurement errors of the variables being studied.
Econometrics combines economics, mathematical economics, and
statistics to give numerical values to the aspects of economic connections.
Econometric methods are used to find the values of parameters that are basically
the coefficients in the mathematical description of economic connections. The
statistical methods that help to explain economic events are modified into
econometric methods. Econometric relationships illustrate the unpredictable
nature of economic connections, which are usually not included in economics
and mathematical models. It is important to note that econometric methods can
also be used in other fields like engineering, biology, medicine, earth sciences,
and agriculture. Simply put, whenever there is a need to find a random
relationship in mathematical terms, econometric methods and tools can assist.
Econometric tools help explain the links between variables. Basically,
econometrics is the use of statistical and mathematical techniques in economics.
E.g.: Analysing the impact of income on household consumption.
The main goal of Mathematical Economics is to represent economic
theory in mathematical forms (equations) without considering whether the
theory can be measured or tested in reality. An econometrician, however, is
particularly interested in testing economic theory through real-world evidence.
As we will see, the econometrician often uses the mathematical equations
created by the mathematical economist but changes them into a format that can
be tested in real life. This transformation from mathematical to econometric
equations requires a lot of creativity and practical skill. Mathematical
economics also looks at how methods from other areas of math can be applied.
For example, Equilibrium Theory uses Fixed-Point Theorems, which depend on
concepts from Analysis and Topology. Mathematical economics applies

5|Page
mathematical methods to represent theories and analyses issues in economics.
Typically, these applied methods go beyond basic geometry and include
differential and integral calculus, difference and differential equations, matrix
algebra, mathematical programming, and other calculation methods. Supporters
of this approach argue that it enables the creation of theoretical links with
precision, broad application, and clarity. Mathematics helps economists develop
clear and testable ideas about diverse and complex topics that are harder to
express informally. Moreover, the language of mathematics lets economists
make specific, positive claims about debated or controversial topics that would
be impossible without it. Much of today’s economic theory is presented using
mathematical economic models, which are a set of simplified mathematical
relationships meant to clarify assumptions and outcomes.
E.g. Comparative statics as to a change from one equilibrium to another induced
by a change in one or more factors.
1.2.3 Types of econometric analysis
Econometrics can be divided into two parts: theoretical and applied
econometrics.
Theoretical econometrics focuses on creating suitable methods for measuring
economic relationships that do not rely on controlled experiments done in labs.
Econometric methods are usually created for studying data that is not collected
through experiments. E.g. Ordinary least square (OLS) estimation and
generalized least square (GLS) estimation.
Applied econometrics involves using econometric methods to address specific
areas of econometric theory and issues like demand, supply, production,
investment, consumption, etc. Applied econometrics uses the tools from
econometric theory to analyse economic events and predict economic
behaviour. E.g. studying the impact of minimum wage increase on employment.
Additionally, econometrics can also be categorised into micro econometrics and
macro econometrics. Micro econometrics deals with the individual level data
such as household level, firm level, and industry level. It analyses economic
behaviors at a smaller [Link] contrast, macro econometrics deals with
aggregate economic indicators like GDP, inflation, and unemployment, helping
to understand large scale economic trend and policy implications.

Self-Assessment Question 1
1. Which of the following is true?
a) Econometrics is the study of economy without data
b) Econometrics is purely mathematical
c) Econometrics test economic theories using relevant data
d) Econometrics is a branch of literature 6|Page
2. The branch of econometrics that deals with developing new estimation
method is called:
1.2.4 Structure of Economic Data

Econometric analysis is done using data. There are different types of economic
data:
 Cross-sectional data: Data collected at a single point in time.
 Time series data: Data collected over different time periods.
 Pooled cross sections: Combination of two or more cross-sectional
datasets.
 Panel/Longitudinal data: Data that follows the same subjects over time.

The method we use in econometrics depends on the type of data. If we use the
wrong method, it may give wrong or misleading results.
1)Cross-Sectional Data
Cross-sectional data is collected at a particular point in time from samples like
individuals, households, firms, cities, or countries. Each observation is mostly
independent of the other. This type of data is often collected using random
sampling. Cross-sectional data is commonly used in applied microeconomics.

2)Time Series Data

Time series data means collecting information about a variable or multiple
variables over time.
Examples include stock prices, money supply, consumer price index, GDP,
automobile sales, etc. These data are usually related over time, which is called
serial correlation. The order of data matters a lot in time series and time series
data can be collected daily, weekly, monthly, quarterly, or yearly. Trends and
seasonality are common features of time series data. It is mostly used in
macroeconomics and finance.

3)Pooled Cross Sections

Pooled cross-section data is when two or more cross-sectional datasets are
combined. Each dataset is collected independently. This method is useful to
study policy changes or changes over time. For example, studying house prices
in 1993 and then again in 1995 to see changes after a policy shift.

4)Panel or Longitudinal Data

Panel data is when we collect data from the same subjects repeatedly over time.
It has both cross-sectional and time series features. Panel data helps us observe
things that do not change over time (time-invariant factors). It is also useful to
study lag effects, where an action shows results after some time. For example,
studying crime rates in a city over two years or seeing how police presence
affects crime over time.
7|Page
1.2.5 Types of economic questions
Econometrics is the statistical toolkit that the economists use to answer
economic questions with data. The types of economic questions which they
analyses are the following;
 Descriptive questions ask about how things are (or were) in reality.
E.g. Has economic inequality increased since 1960?
Here we only observe data for a sample of individuals, not for the full
population. The sample is selected randomly from the population. Let us
imagine we want to know how the distribution of income in India has
changed. But we only observe for a survey of workers.
 Causal question asks about what would have happened dealing with the
cause-and-effect relationships between economic variables.
E.g. How do increases in minimum wage affect the employment?

Answering causal questions are harder than descriptive ones. Because the
casual questions involve both descriptive component (what are the
outcomes in reality) and counterfactual component (how would things
have been under a different treatment?). Counterfactual question cannot
ever be answer with data alone as it need additional assumptions to learn
about more.
Selection bias: Refers to the settings where the sample is not drawn randomly
from the population. Thus, sample does not represent population leading to
misleading results.
1.2.6 Framework for approaching econometric analysis
Suppose an economist is examining two educational institutions, A and B, and
aims to determine the causal effect of attending Institution A on earnings
relative to Institution B.
Population: population refers to the entire group of individuals, objects, or
observations that share a common characteristic.
E.g. In this context, the population refers to all individuals who could
potentially attend either Institution A or Institution B. This could include all
students eligible for admission to both institutions or a specific subgroup, such
as graduates from both institutions.
Sample: the data that the economist actually observes as studying about the
entire population is impractical. Sample is the subset of population.
E.g. A survey of students from two different educational institutions (A and B)
about their earnings after graduation.

8|Page
Estimator: a function of the data in the sample.
E.g. Difference in earnings between students from the two educational
institutions in the survey.
Estimand: a function of the observable data for the population.
E.g. Difference in earnings between all students from the two educational
institutions.
Target parameter: it is a numerical value that describes a characteristic of a
population that is to be tested or estimated. A parameter is a fixed value, but in
practice, it is often unknown because it is difficult to measure an entire
population.
E.g. Causal effect on earnings of attending Institution A relative to Institution B
Statistic – A statistic is a numerical value that describes a characteristic of a
sample (a subset of the population). It is used to estimate the corresponding
population parameter.
The process of learning about the estimand from the estimator constructed with
your sample is called statistical estimation/inference. The process of learning
about the parameter from the estimand is called identification.

Self-Assessment Question 2
1. Data collected at a single point of time from multiple entities is called
as:
a) Panel data
b) Cross-sectional data
c) Time series data
d) Pooled cross data
2. What is an Estimand?
a) A statistical method
b) A prediction
c) A sample statistic
d) A population parameter
3. Which of the following is an example of an estimator
a) Population mean
b) Sample mean
c) Regression coefficient
d) All of the above
4. True or false
I. A statistic is the measure that describes the entire population
II. Selection bias is a statistical error that occurs when the
sample collected is the exact representation of the population.

1.3 SCOPE OF ECONOMETRICS

9|Page
As mentioned earlier, econometrics applies statistical and mathematical
methods to analyse economic data, test hypotheses, and estimate relationships.
In contrast, mathematical economics focuses on expressing economic theories
and relationships using mathematical tools such as calculus, algebra, and
optimization. It develops precise theoretical models but does not test them
against real-world data. Mathematical economics builds models whereas
econometrics verifies their validity using empirical data. Statistics provides
general tools for analysing data, but in econometrics, we apply these methods to
economic issues. Thus, econometrics is studied separately as it bridges the gap
between theory and real-world economic analysis, making it essential for policy
evaluation and forecasting.
Goals/Scope of Econometrics refers to the importance or usefulness of the field
of Econometrics. There are mainly four main goals for the subject. They are,
 Analysis: Testing economic theory and hypotheses
 Evaluating and implementing business and government policies
 Forecasting economic variables
 Estimating relationships between economic variables
1) Analysis: Testing Economic Theory and Hypotheses
In the beginning, economic theories were based on a set of belief or
assumptions about how individual units (like consumers, producers, etc.)
behave. From these assumptions, economists made some general conclusions or
rules that describe how the economy functions. These economic theories were
not tested against real economic situations. In other words, there were no
attempts to see if the theories truly explained how individuals act in the
economy. Econometrics mainly focuses on verifying those economic theories.
In this context, we can say that the goal of the research is 'analysis', which
means gathering evidence (data) to test how well economic theories explain the
actual behaviour of economic units. Today, any economic theory, no matter how
well it is presented or logically sound, cannot be accepted without testing with
real data.
2) Evaluating and implementing business and government policies
By evaluating we aim to get numerical estimates of the coefficients of economic
relationships for policy simulations. In many situations, we use different
econometric methods to obtain reliable values of the individual coefficients of
economic relationships. This allows us to assess the major aspects of economic
theory. Understanding how to get these numerical values (for example, ceteris
paribus, marginal concepts in economics, the concept of multiplier, technical
coefficients of production, etc.) is very important for creating business and
government policies. It helps to compare the outcomes of different policy
choices. For example, a government's decision to increase public spending to
10 | P a g e
boost economic growth depends on the numerical values of the fiscal multiplier
and the marginal propensity to consume (MPC). If the fiscal multiplier is low,
higher government spending may not significantly stimulate the economy. This
highlights the importance of accurately estimating economic coefficients.
Econometrics helps determine these values, making it an essential tool for
designing effective fiscal policies. Econometrics can provide these numerical
estimates and has become a key tool for creating effective economic policies.
3) Forecasting economic variables
Predicting the values of economic variables is essential for developing various
economic policies, and econometrics is very helpful for this forecasting and
policy development. For instance, if the government is considering an
unemployment policy, it is crucial to estimate the unemployment rate over the
next five years without intervention. These projections help policymakers
respond effectively to different scenarios:
 If future unemployment remains low, the government can focus on
maintaining economic stability.
 If unemployment is expected to rise, proactive measures such as job
creation programs or skill development initiatives should be
implemented.
Forecasting is thus becoming more important for managing developed
economies as well as for planning economic growth for underdeveloped
countries.
4) Estimating relationships between economic variables
By using regression analysis, testing hypotheses, and analysing time series,
economists can measure the connections between economic variables, recognize
the elements that affect economic results, and assess how well economic
policies work. For instance, we can examine how a rise in unemployment
impacts the inflation rate in an economy, allowing policymakers to adjust their
strategies accordingly.
1.3.1 Examples of research problems
 Modelling the long-term relationships between prices and interest rates
 Looking at how inflation impacts unemployment rates
 Looking at how disposable income influences consumer spending
 Identifying the factors that impact GDP per capita growth
 Predicting the relationship between returns and the stock market indices
of two countries

11 | P a g e
1.3.2 Application of econometrics in specific fields
 Finance and Investment
- Estimating stock price movements and risk factors.
- Analysing the impact of interest rates on investment returns.
 Labor Economics
- Examining the relationship between wages and employment levels.
- Assessing the effects of education on earnings.
 Macroeconomics
- Forecasting GDP growth and inflation trends.
- Evaluating the impact of fiscal and monetary policies.
 Public Policy and Governance
- Measuring the effectiveness of social welfare programs.
- Predicting the impact of tax changes on income distribution.
 International Trade
- Analysing the effects of trade policies on exports and imports.
- Estimating exchange rate fluctuations and their economic impact.
Self-Assessment Question 3

1. Which of the following is NOT a key application of econometrics?

a) Forecasting economic trends
b) Estimating economic relationships
c) Developing political ideologies
d) Evaluating government policies
2. How does econometrics contribute to economic forecasting?
a) By using historical data and statistical models to predict future trends
b) By speculating future trends without using data
c) By assuming economic behavior remains unchanged over time
d) By avoiding the use of mathematical equations

1.4 METHODOLOGY OF ECONOMETRIC RESEARCH

How does an econometrician analyse an economic theory? This is where the
methodology of econometrics comes in, which is a step-by-step process. These
steps are:
1. Statement of the theory/hypothesis
A theory should make a prediction. In statistics and econometrics, we also talk
about a hypothesis. A hypothesis is an if-then statement which is tentative, and
the theory is essentially a verified hypothesis. For example, the Phillips Curve
suggests that if unemployment decreases, then inflation increases. Another
12 | P a g e
example is the Quantity Theory of Money, which states that if the money supply
increases significantly, then inflation will rise in the long run. Similarly, the
hypothesis that if interest rates decrease, then business investments will
increase, can be tested and potentially verified as a theory.
2. Specification of Mathematical Model
A model is a simplified representation of reality. It should be representative,
meaning it should include the important features of the situation being studied.
Generally, one of the goals in modelling is to create a simple model to explain a
complex issue. This goal can sometimes lead to overly simple models, and
sometimes the assumptions can be unrealistic. This is where algebra comes in.
We need to use mathematical skills to create an equation. For example, assume
a theory that says more years of schooling leads to higher wages, meaning there
is a positive link between years of schooling and wage rate. Economic model of
years of schooling and wage rate of individuals is formulated. The formal
economic theory not really needed to derive equation.
Wage= f (years of education)
In economic terms, we say the return on schooling is positive for wages. The
equation is: Y = β0 + β1X. Where; Y is the dependent variable representing
wages, β0 is a constant, and β1 is the coefficient for years of schooling. X is the
independent variable that measures schooling, meaning the number of years
spent in school. We also refer to β0 as the intercept and β1 as the slope
coefficient. Usually, we would expect both β0 and β1 to be positive.
3. Specification of Econometric Model
An economic model is a collection of assumptions that describes how an
economy works or more generally, a phenomenon. In practice, all the variables
that the researcher thinks are important in explaining the phenomenon are
included in the model. The other variables are grouped into a category called
“disturbances,” where disturbances are random variables whose behaviour is
unpredictable. This is the main difference between economic modelling and
econometric modelling. This is also the main difference between mathematical
modelling and statistical modelling. Mathematical modelling is exact, while
econometric modelling includes a random term as well. Here, we believe that
the mathematical model is correct, but we should consider that it might not be.
We add an error term, u, to the equation above. This is also known as a random
variable or stochastic variable. It stands for other factors that we cannot measure
or do not know that influence Y. It also accounts for measurement errors that
might have affected the data. The econometric equation is: Y = β 0 + β1X + u.
The error term, U, is believed to follow a certain type of statistical distribution.
4. Collection of Data
13 | P a g e
We need data for the variables above. This can be gathered from government
statistics offices and other sources. Many data sets can also be found online
these days. However, we need to learn how to locate the right data among the
huge amounts of information available. Unlike physical sciences, much of the
data gathered in economics (like GDP, money supply, Dow Jones index, car
sales) is nonexperimental (observational) because the organization collecting
the data such as the government may not have any direct control over it. Thus,
the information on years of schooling and wage rates comes from details
provided to the government by individuals in the labor market. In a way, the
government is a passive collector of this information and may not account for
theories such as human capital theory or signalling theory, or any other relevant
frameworks. As a result, the gathered data may reflect various factors
influencing an individual's earnings beyond just education, such as skills,
experience, and labor market conditions. In other words, the same data might
support more than one theory regarding the relationship between schooling and
wages. Different types of data are used to estimate the model are time series
data, cross-section data, pooled cross data, and panel data.
5. Estimation of the econometric model
Here, we measure β0 and β1, which means we get numerical values for the
targeted parameters. This is done using a statistical method called regression
analysis. This gives real support to the theory we are looking at.
6. Hypothesis Testing
Hypothesis testing is the statistical technique used for making decisions or draw
conclusions about a population based on sample data. In this step of testing the
hypothesis, we first need to think about the theory and the hypothesis we
explained earlier. Our theory predicted that years of schooling is positively
related with wages. Does the econometric model support this theory? What we
do here is called statistical inference (hypothesis testing). To show this positive
link between years of education and wages, the β1 coefficient of the econometric
model should be above 0.
7. Forecasting/prediction
If the hypothesis testing was positive, meaning the theory is deemed correct, we
predict the wage values by estimating the values of education. For instance, how
much would someone earn for one more year of schooling? If the X variable is
the years of schooling, the β1 coefficient provides the answer to this question.
8. Use of the model for policy purpose
Finally, if the theory seems logical and the econometric model was not
disproven based on the hypothesis test, we can use the theory for policy
suggestions.
14 | P a g e
How to choose between different competing models of econometrics or
explanations?
When a student encounters a new economic theory or model, he/she might
identify answers to the questions: first and foremost, the purpose of the theory
and how does it influence economic choices and preferences. Then one should
look into the proof or evidence available which allows us to assess the quality
compared to other theories or models. The eight-step classical econometric
method mentioned earlier is unbiased because it can be used to examine any of
these competing ideas.
1.4.1 Assumptions of econometric analysis
 Relationships between economic variables are quantifiable. That is, these
economic relationships can be expressed in mathematical or statistical
form.
 Assumes that sufficient and reliable data is available for econometric
modelling. The data available is also relevant.
 Randomness and uncertainty exist and the real-world data is affected by
random factors which is represented as disturbance or error term. This is
accountable in the analysis.
 Appropriate econometric models can be specified based on verified
economic theories. That is, model specification is possible.
 Econometrics assumes that causal relationships can be tested not just
correlations.
 An objective, systematic and scientific method I used to analyse
economic problems
1.4.2 Steps in econometric analysis

15 | P a g e
Fig 1.2 Econometric analysis starts with economic theory, which helps build an econometric
model using relevant data. The model is then estimated to find relationships between
variables. After estimation, specification tests and diagnostics check if the model is
appropriate. If the model is adequate, hypothesis testing is done to verify economic
predictions. A valid model is finally used for forecasting and policymaking.

Self-Assessment
The process Question 4
of econometric analysis starts with economic theory, which
acts 1.
as the
What base forfirst
is the creating
step inaneconometric
econometric model. The first step is always
analysis?
crucial a)asData
it involves
collectionconverting verified economic theory into precise
econometric model. This
b) Hypothesis model is then backed by data, which is gathered to
testing
estimatec)the connectionsofand
Specification the relationships
model outlined in the model (Here notice
that the steps in econometric
d) Interpretation analysis and steps in econometric modelling (OLS)
of results
are different). Next, the model
2. In econometrics, estimation
hypothesis testingphase uses
is used to: statistical methods to find
the numerical
a) Collectvalues
dataof the model’s parameters. After the estimation, the model
goes through specification
b) Validate economic testing and diagnostics to check if it is accurate and
theories
trustworthy. At this
c) Avoid point, the
statistical question that arises is stated: Is the model good
inference
enough?d)IfIgnore
it is relationships
not, changes between
are required.
variablesThus, the process goes back to
adjusting
3. Thetheprocess
theory or of the data. If the
estimating the parameters
model is good of anenough, it moves
econometric on to
model is the
hypothesis testing phase, where economic ideas are examined using the
called:
estimated connections. Finally, once verified, the model is used for making
a) Specification
predictions and decisions,
b) Hypothesis testinggiving valuable information for economic policy
choices c)
andEstimation
planning. Although we explored econometric methodology through
an example from labour economics earlier, the same approach can be applied to
d) Prediction
studying
4. Ifquantitative
an estimatedrelationships
model does not between
fit thevariables
data well,inthe
various fields.
next step In fact,
should
regression
be:analysis has been utilized in disciplines such as politics, international
relations,a)psychology,
Refining thesociology,
model by meteorology, and many others.
respecifying variables
 Anb) Accepting
objective, the model as and
systematic it is scientific method is used to analyse
c) Ignoring
economic errors
problems.
d) Avoiding statistical correction

16 | P a g e
1.5 LIMITATIONS OF ECONOMETRICS
Econometrics, like any other subject, has certain limitations that affect its
Self-Assessment
applicability. SomeQuestions 5 discussed in this section. One major limitation is
of them are
[Link]
focuses only
of the on quantifiable
following economic ofvariables
is a major limitation such as prices,
econometrics?
production, and employment
a) It cannot be applied to etc. Sometimes
economic it ignores
policy analysisqualitative aspects like
consumerb) Itsatisfaction,
completely innovation, and policy in
eliminates uncertainty effectiveness, which are harder to
economic predictions
quantify.
c) ItAnother
relies onissue is the dependence
assumptions that may notonalways
data availability
hold true in and accuracy.
real-world
Econometric
scenariosanalysis assumes that sufficient and reliable data is available for
econometric modelling
d) It does anddata
not require the for
data available is also relevant. Econometric
analysis
models require large and reliable datasets, but in many cases, data may be
incomplete, outdated, or even misleading, which can lead to incorrect
conclusions.
Additionally, sampling methods introduce errors in econometric
predictions. Since most econometric studies are based on sample data rather
than the entire population, they inherit the limitations of sampling techniques,
such as selection bias and measurement errors. Moreover, statistical
assumptions in econometrics may not always hold in real-world data. Many
econometric models assume linear relationships, normality in error terms, or
absence of multicollinearity, but economic data is often complex and does not
always follow these ideal conditions. Lastly, econometric methods are time-
consuming and complex, requiring a deep understanding of mathematics,
statistics, and computing. Even minor mistakes in model specification or data
handling can lead to significant errors in results, making econometrics a
challenging field to apply accurately. Despite these limitations, econometrics

17 | P a g e
remains a powerful tool for analysing economic relationships and guiding
policy decisions.

1.6 FUTURE RELEVANCE AND EMERGING AREAS IN

ECONOMETRICS
Big Data and Machine Learning Integration: The growing availability of large
data sets has resulted in the use of machine learning methods in economic
modelling. Techniques like deep learning, neural networks, and combined
models are enhancing forecasts and uncovering complex economic connections.

Behavioural and Experimental Econometrics: Standard econometrics assumes

logical decision-making, while behavioral econometrics includes psychological
factors to better understand how consumers and investors act. Experimental
econometrics uses controlled tests to evaluate economic theories more
effectively.

Financial Econometrics and Risk Modelling: The rising complexity of financial

markets has increased the demand for advanced econometric models in risk
evaluation, asset pricing, and portfolio management. Sophisticated time series
models assist in predicting stock market changes, interest rates, and currency
rate shifts.
Climate and Environmental Econometrics: With rising worries about climate
change, econometric models are being used to evaluate the economic effects of
environmental policies, carbon taxation, and sustainable development plans.

Real-Time and High-Frequency Data Analysis: The growth of high-frequency

trading and real-time economic tracking has created the need for models that
can quickly process and analyse data, aiding policymakers and businesses in
making fast and informed choices.

Health and Pandemic Econometrics: The COVID-19 pandemic emphasized the

significance of econometrics in healthcare. It is used to evaluate the economic
effects of health policies, forecast disease spread, and improve resource
distribution.

Policy Evaluation and Causal Inference: Advanced econometric methods, such

as Difference-in-Differences (DiD) and Instrumental Variable (IV) techniques,
are increasingly utilized to assess the effectiveness of government policies and
social programs.

18 | P a g e
Cryptocurrency and Blockchain Econometrics: With the rise of digital money,
econometric models are being created to analyse cryptocurrency markets,
blockchain transactions, and the effects of decentralized finance (DeFi).
As the world becomes more focused on data, econometrics will keep growing
its uses in various fields, making it a vital tool for economic analysis and
decision-making.

1.7 CONCLUSION
Econometrics is crucial for understanding and exploring economic relationships
using mathematical and statistical methods. While it provides helpful insights in
creating policies (both government and business policies), its effectiveness
depends on having quality data and the accuracy of the underlying theories.
Despite some limitations, econometrics continues to be an essential tool for
economists, researchers, and policymakers. With ongoing advancements in
technology and data analysis, its applications are increasing, making it even
more vital for addressing complicated economic problems in the future.

1.8 SUMMARY
This chapter explored the nature, scope, and methodologies of econometrics in
detail. The first section provided the definition and meaning of econometrics,
explored its different types and mentioned relevant examples for each. It also
discussed how economic theories, mathematical economics and econometrics
are connected, pointing out the differences between mathematical economics
and econometrics as separate fields.
The next section expanded on the scope and goals of econometrics,
stressing its position as a separate area from mathematics and statistics. This
discussion also examined the key applications of econometrics across different
branches of economics. The third section focused on econometric
methodologies, outlining the essential steps involved in the modelling process,
supported by relevant economic theories and practical examples.
Finally, the chapter talked about the limits of econometrics, recognizing
issues such as data limitations and dependence on assumptions. It also looked at
the future of econometrics, emphasizing its changing role in economic research
and policy-making. The key points are
o Econometrics: econometrics is the science and art of using economic data
to measure the empirical validity of economic theory by applying
mathematical and statistical tools.
o Differences between mathematical economics and econometrics: the
primary focus of mathematical economics is to represent economic theory
using mathematical equations or models, without considering their
19 | P a g e
 measurability or empirical verification. In contrast, econometrics is
mainly concerned with empirically verifying economic theories. As will
be demonstrated, econometricians often utilize mathematical models
developed by mathematical economists but modify them into forms that
allow for empirical testing.
o Types of econometrics: econometrics can be broadly classified into
theoretical econometrics and applied econometrics. Theoretical
econometrics focuses on developing new statistical methods and refining
existing techniques for analysing economic data. Applied econometrics,
on the other hand, involves using these statistical methods to study real-
world economic problems, test economic theories, and aid in policy-
making.
o Types of economic questions: descriptive questions ask about how things
are (or were) in reality whereas causal question asks about what would
have happened
o Sample: a sample is a subset of a population selected for analysis,
o Population: a population refers to the entire group of individuals, units, or
observations that share a common characteristic and are the focus of a
study.
o The process of learning about the estimand from the estimator
constructed with your sample is called statistical estimation/inference.
The process of learning about the parameter from the estimand is called
identification
o Goals of econometrics: goals/scope of econometrics refers to the
importance or usefulness of the field of econometrics. There are mainly
four main goals for the subject. They are, testing economic theory and
hypotheses, evaluating and implementing business and government
policies, forecasting economic variables, and estimating relationships
between economic variables.
o Methodology: methodologies include stating the theory/hypothesis,
specification of mathematical model, specification of econometric model,
collection of data, estimation of the econometric model, hypothesis
testing, forecasting/prediction, and use of the model for policy purpose.
o Limitations of econometric analysis: econometrics is limited by the
availability and accuracy of data, as unreliable or insufficient data can
lead to misleading results. Additionally, econometric models rely on
assumptions that may not always hold in real-world economic conditions,
affecting their predictive accuracy.

1.9 GLOSSARY
 Econometrics – A branch of economics that applies statistical and
mathematical methods to analyse economic data and test theories.
20 | P a g e
 Economic Theory – The conceptual framework that explains economic
behaviors and relationships, forming the basis for econometric analysis.

 Mathematical Economics – The use of mathematical expressions and models

to represent economic theories, without necessarily considering their
empirical verification.

 Statistical Methods – Techniques used in econometrics, such as probability

distributions, hypothesis testing, and regression analysis, to analyse data and
draw conclusions.

 Empirical Analysis – The process of using real-world data to test economic

theories and validate models through statistical evaluation.

 Regression Analysis – A key econometric tool that examines the relationship

between dependent and independent variables to estimate economic
relationships.

 Econometric Modelling – The formulation of mathematical models that

represent economic relationships and can be tested using statistical methods.

 Hypothesis Testing – A statistical technique used to test assumptions or

predictions in econometrics, helping to confirm or reject economic theories.

 Data Estimation – The process of deriving numerical values for economic

relationships using econometric techniques like Ordinary Least Squares
(OLS).

 Forecasting – The use of econometric models to predict future economic

trends based on historical data and statistical techniques.

 Policy Evaluation – Assessing the impact of government policies (e.g.,

taxation, subsidies, or trade policies) using econometric methods to
determine their effectiveness.

 Micro econometrics – A subfield of econometrics that focuses on individual-

level data, such as households, firms, or industries, to analyse economic
behaviours.

21 | P a g e
 Macro econometrics – The application of econometric methods to aggregate
economic data, such as GDP, inflation, and unemployment, to study large-
scale economic trends.

 Limitations of Econometrics – The challenges faced in econometric analysis,

such as data availability, model assumptions, and potential errors in
estimation.

 Economic Relationships – The interconnections between different economic

variables, such as supply and demand, consumption and income, or inflation
and unemployment.

 Quantitative Analysis – The use of mathematical and statistical tools to

measure and analyse economic phenomena.

 Predictive Modelling – Building models that estimate future outcomes based

on historical data and econometric techniques.

 Time Series Analysis – A statistical approach in econometrics that examines

data points collected over time to identify trends, patterns, and seasonal
effects.

 Panel Data Analysis – A method of econometric analysis that combines

cross-sectional and time-series data, allowing for the study of economic
behavior across multiple entities over time.

 Cross-Sectional Data – Data collected at a single point in time across

different individuals, firms, or countries, often used for comparative
economic studies

1.10 TERMINAL QUESTIONS

Short Answer Type Questions
1. Define econometrics.
2. What are the key components of econometric analysis?
3. What is the primary objective of econometrics?
4. List the general steps in the econometric methodology.
5. Why is hypothesis testing important in econometrics?
6. What is the role of model specification in econometrics?
7. How does econometrics assist in economic policy evaluation?
8. Why is data collection crucial in econometric analysis?
9. How do descriptive economic questions help in policy-making?
22 | P a g e
 [Link] is the role of causal economic questions in econometrics?
[Link] are forecasting economic questions important in decision-making?
[Link] between theoretical and applied econometrics.
[Link] is the focus of micro econometrics?
[Link] are the major types of economic data?
[Link] time series data with an example.
[Link] does cross-sectional data differ from time series data?
[Link] is panel data, and why is it useful in econometrics?
[Link] is the choice of data type important in econometric analysis?
[Link] does economic theory relate to econometrics?
[Link] is a common limitation of econometric models?
Long answer type question
1. Differentiate between mathematical economics and econometrics? Why
do we study econometrics as a different discipline from mathematical
economics and statistics?
2. What are the different scopes of econometrics? Explain each.
3. The table shows the relationship between minimum wages, employment
growth, and unemployment rates in India over the years. Explain the
economic theory behind the impact on minimum wages on employment
and identify the different steps involved in the econometric modelling.

Year Minimum Wage (₹ Employment Growth Unemployment Rate

per day) (%) (%)
1990 35 2.5 2.4
2000 52 2.3 3.1
2010 100 1.9 3.8
2020 176 1.4 6.1
2023 225 1.2 7.8
Source: Ministry of Labour & Employment, CMIE

1.11 KEY ANSWERS

Self-Assessment Type Questions

(1)
1. c) Econometrics test economic theories using relevant data 2. b)
Theoretical econometrics 3. False. Econometrics uses observational data
opposed to experimental data to measure the economic relationships.
(2)

23 | P a g e
 1. b) Cross-sectional data 2. d) A population parameter 3. b) Sample mean
4. False, False
(3)
1. c) Developing political ideologies 2. a) By using historical data and
statistical models to predict future trends
(4)
1. c) Specification of the model 2. b) Validate economic theories 3. c)
Estimation 4. a) Refining the model by respecifying variables
(5)
1. It relies on assumptions that may not always hold true in real-world
scenarios

Short Answer Type Questions

1. Econometrics applies statistical and mathematical methods to analyse
economic data and test economic theories.
2. Economic theory, data collection, model specification, estimation,
hypothesis testing, and interpretation.
3. To quantify economic relationships, test hypotheses, and make forecasts
based on empirical data.
4. Problem definition, model specification, data collection, estimation,
hypothesis testing, and interpretation.
5. It helps determine the validity of economic relationships and assess model
accuracy.
6. It defines the functional form of the relationship between variables and
ensures correct estimation.
7. By estimating policy impacts, predicting outcomes, and guiding evidence-
based decision-making.
8. Reliable data ensures accurate model estimation and meaningful
conclusions.
9. They summarize economic conditions and trends, aiding policymakers in
informed decisions.
[Link] identify cause-and-effect relationships to understand policy impacts.
[Link] help predict future trends and guide businesses and policymakers.
[Link] econometrics develops methods, while applied econometrics
uses them on real-world data.
[Link] studies individual and firm-level economic behavior using microdata.
[Link] series, cross-sectional, pooled cross and panel data.
[Link] collected over time, e.g., India's GDP from 1990 to 2023.
24 | P a g e
[Link]-sectional data captures multiple observations at a single point in time,
unlike time series, which tracks changes over time.
[Link] data combines cross-sectional and time-series data, allowing for better
control of individual differences.
[Link] determines the appropriate model and estimation techniques.
[Link] provides the foundation for model formulation and hypothesis testing.
[Link] rely on assumptions that may not always hold in real-world scenarios.

Long Answer Type Questions

1. Mathematical economics focuses on expressing economic theories and
relationships using mathematical tools such as calculus, algebra, and
optimization. It develops precise theoretical models but does not test them
against real-world data. In contrast, econometrics applies statistical methods
to analyse economic data, test hypotheses, and estimate relationships. While
mathematical economics builds models, econometrics verifies their validity
using empirical data. Unlike statistics, which provides general tools for data
analysis, econometrics specifically applies these techniques to economic
problems. Thus, econometrics is studied separately as it bridges the gap
between theory and real-world economic analysis, making it essential for
policy evaluation and forecasting.
2. Econometrics has several key areas of application. Theoretical econometrics
focuses on developing new statistical methods for economic data analysis,
ensuring model assumptions and estimation techniques are valid. Applied
econometrics uses these models to analyse real-world economic issues, such
as the impact of inflation on unemployment. Micro econometrics deals with
individual, household, or firm-level data to study behaviors like consumer
demand, while macro econometrics examines economy-wide issues such as
GDP growth, inflation, and monetary policy. Another important field is
financial econometrics, which applies econometric methods to financial
markets, including modelling stock price volatility and risk assessment.
3. The economic theory behind minimum wages and employment suggests that
a higher minimum wage raises labour costs for firms, potentially leading to
reduced hiring, especially in industries with elastic labour demand. If firms
cannot afford to pay higher wages, they may cut jobs, reduce working hours,
or shift towards automation. In countries like India, where the informal
sector is large, higher minimum wages may push workers into informal jobs
where wage regulations are not strictly enforced. To analyse this impact
using econometrics, the first step is data collection, which involves gathering
information on minimum wages, employment rates, and control variables
like inflation. The next step is model specification, where an equation is
25 | P a g e
 formulated to express employment as a function of minimum wages and
other factors. The model is then estimated using methods such as Ordinary
Least Squares (OLS) regression or Fixed Effects models. Hypothesis testing
is performed to determine whether changes in minimum wages significantly
impact employment. Finally, the results are interpreted to assess whether
wage increases lead to job losses, higher productivity, or shifts to informal
employment.

1.12 REFERENCES
1. Damodar N Gujarati and Dawn C Porter (2009): Basic Econometrics,
Fifth Edition, McGraw Hill International Edition.

2. Damodar N Gujarati
(2011): Econometrics
by Example,
3. First Edition,
Palgrave, MacMillan.
2. Damodar N Gujarati (2011): Econometrics by Example, First Edition,
Palgrave, MacMillan.
3. Jeffrey M Wooldridge (2018): Introductory Econometrics: A Modern
Approach, 7th Edition, Thomson South Western.
4. Koutsoyiannis A (2001). Theory of Econometrics, Palgrave
Macmillan, 2nd Edition.
5. Maddala G.S. (2010). An Introduction to Econometrics, Whiley, 3rd
Edition.
6. Kennedy P (2004), A Guide to Econometrics, Blackwell Publishers, 5th
Edition.
7. James H Stock and Mark W Watson (2017): Introduction to
Econometrics, Third Edition, Pearson, Addison Wesley.
8. Kerry Patterson (2000): An introduction to Applied Econometrics: A

Time Series Approach, First Edition, Palgrave.

9. Walter Enders (2010): Applied Econometric Time Series, Third Edition,
Wiley India Edition.
26 | P a g e
10. Richard Harris and Robert Sollis (2006): Applied Time Series Modelling
and Forecasting, First Edition, Wiley Student Edition.
11. Dimitrios Asteroid and Robert Hall (2015): Applied Econometrics, 3rd
Edition, Oxford University Press.
12. Jack Johnston and John DiNardo (1998): Econometrics Methods, Fourth
Edition, The McGraw Hill Companies.
13. William H Greene (2018): Econometric Analysis, 8th Edition, Pearson
Education.
14. Christopher Dougherty (2007): Introduction to Econometrics, Third
Edition, Oxford University Press.
15. Chris Brooks (2012): Introductory Econometrics for Finance, 3rd
Edition, Cambridge.

16. Walter Enders

(2010): Applied
Econometric Time
Series,
17. Third Edition,
Wiley India Edition
18. Kerry Patterson
(2000): An
introduction to Applied
19. Econometrics: A
Time Series Approach,
First Edition,
27 | P a g e
20. Palgrave.
21. Kerry Patterson
(2000): An
introduction to Applied
22. Econometrics: A
Time Series Approach,
First Edition,
23. Palgrave.
24. 8. Walter Enders
(2010): Applied
Econometric Time
Series,
25. Third Edition,
Wiley India Edition.
26. 9. Richard Harris
and Robert Sollis
(2006): Applied Time
28 | P a g e
27. Series Modelling
and Forecasting, First
Edition, Wiley
28. Student Edition.
29. 10. Dimitrios
Asteroid and Robert
Hall (2015): Applied
30. Econometrics, 3rd
Edition, Oxford
University Press.
31. 11. Jack Johnston
and John DiNardo
(1998): Econometrics
32. Methods, Fourth
Edition, The McGraw
Hill Companies.

29 | P a g e
33. 12. William H
Greene (2018):
Econometric Analysis,
8th
34. Edition, Pearson
Education.
35. Kerry Patterson
(2000): An
introduction to Applied
36. Econometrics: A
Time Series Approach,
First Edition,
37. Palgrave.
38. 8. Walter Enders
(2010): Applied
Econometric Time
Series,
30 | P a g e
39. Third Edition,
Wiley India Edition.
40. 9. Richard Harris
and Robert Sollis
(2006): Applied Time
41. Series Modelling
and Forecasting, First
Edition, Wiley
42. Student Edition.
43. 10. Dimitrios
Asteroid and Robert
Hall (2015): Applied
44. Econometrics, 3rd
Edition, Oxford
University Press.

31 | P a g e
45. 11. Jack Johnston
and John DiNardo
(1998): Econometrics
46. Methods, Fourth
Edition, The McGraw
Hill Companies.
47. 12. William H
Greene (2018):
Econometric Analysis,
8th
48. Edition, Pearson
Education.
49. ccJames H Stock
and Mark W Watson
(2017): Introduction

32 | P a g e
50. to Econometrics,
Third Edition, Pearson,
Addison Wesley.
51. James H Stock and
Mark W Watson
(2017): Introduction
52. to Econometrics,
Third Edition, Pearson,
Addison Wesley

33 | P a g e