Introduction to Econometrics

Fourth Edition, Global Edition

Chapter 1
Economic questions and
data

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Introduction
• Economics suggests important relationships:
– What is the effect of an increase in the price of cigarettes on
its consumption?
– What is the effect of reducing class size on student achievement?
– How does another year of education change earnings?
• Economic theory provides clues about the answer...
– Cigarette consumption ought to go down when its price goes up
• ...But never suggests quantitative magnitudes of causal effects:
– The actual value of the number (of cigarettes) must be learned
empirically, that is, by analyzing data
Our answers always have some uncertainty: a different set of data would
produce a different numerical answer, because the price is not the only variable
that affects the consumption (smoker's income, potential smokers...)
Copyright © 2019 Pearson Education Ltd. All Rights Reserved.
Introduction

• Econometrics is:
– The science of testing economic theories
– The set of tools used for forecasting (predicting) future
values of economic variables (firm's sales, stock prices...)
– The process of fitting mathematical economic models to real-
world data
– The art of using historical data to make
numerical/quantitative policy recommendations in
government and business

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Introduction
• At a broad level, Econometrics is the science and art of using
economic theory and statistical techniques to analyze economic
data

• Econometric methods are used in many branches of economics

(macroeconomics, microeconomics, economic policy...), in
finance, in marketing, in political science and in other social
sciences

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Formulation of Econometric Models

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Estimation of causal effects
How best might we measure the causal effect on tomato yield of
applying a certain amount of fertilizer?
The best way is to conduct an experiment: In that experiment, a
horticultural researcher plants many plots of tomatoes. Each plot is
tended identically, with one exception: Some plots get fertilizer,
while the rest don't. Whether or not a plot is fertilized is
determined randomly by a computer, ensuring that any other
differences between the plots are unrelated to whether they receive
fertilizer.
This is an example of a randomized (the treatment is assigned
randomly) controlled (control group) experiment

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Estimation of causal effects
• Experiments are used increasingly widely in econometrics
• But, in many applications experiments are not an option because
they are:
– Unethical
– Impossible to execute satisfactorily
– Too time-consuming
– Prohibitively expensive
• Instead, most economic data are obtained by observing real-
world behavior (monetary policy, cigarette prices...)

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Estimation of causal effects
• Even with non-experimental data, the concept of an ideal
randomized controlled experiment is important because it
provides a definition of a causal effect
• In such an experiment, the only systematic reason for
differences in outcomes between the treatment and control
groups is the treatment itself

Causal effect is defined to be the effect of a given action or

treatment on an outcome

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Data sources
• Data come from one of two sources:
1. Experimental data come from experiments designed to evaluate a
treatment or policy or to investigate a causal effect
2. Observational data are collected using surveys, such as telephone
surveys of consumers, administrative records...

• Observational data pose major challenges to econometric

attempts to estimate causal effects, and the tools of econometrics
are designed to tackle these challenges:
– Confounding effects (omitted factors)
– Simultaneous causality
– “Correlation does not imply causation”

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Data sources
• Omitted factors: Factors that influences the outcome other than
the action or the treatment and which are not taken into
consideration
Example: What is the effect of reducing class size on student
achievement?
Student achievement (outcome) is not only influenced by the class
size (action or treatment)
Factors such as teacher quality, family background, computer
usage… can also influence student achievement → omitted factors

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Data sources
• Simultaneous causality: When causality runs “backward” as
well as “forward”
Example: The U.S. is an important trading partner of the E.U.
How sensitive E.U. exports are to fluctuations in US income ?
There is a simultaneous causality in this relationship:
- A decline in U.S. income reduces the E.U. exports
- A decline in E.U. exports reduces E.U. income, which reduces
demand for imports from the U.S., which reduces U.S. income

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Data sources
• Correlation does not imply causation: Just because two
variables are correlated (moving together) does not necessarily
mean that one causes the other
Example: A researcher collected data on ice cream sales and
homicides in New York and found that the two are correlated
Does the consumption of ice cream is causing the death of the
people (or vice versa) ?
No, correlation doesn’t mean causality in our example
- In some cases, correlation occurs by coincidence
- In some other cases, there are some hidden factors which are related on
some level (hot weather in our example?)

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Data types
• Data sets come in three main types:
1. Cross-sectional data consist of multiple entities observed at a
single time period
2. Time series data consist of a single entity observed at multiple
time periods
3. Panel data (also known as longitudinal data) consist of
multiple entities, where each entity is observed at two or more
time periods

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Data types
Example of cross-sectional data

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Data types
Example of time series data

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

Data types
Example of panel data

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.

In this course you will:
• Learn methods for estimating causal effects using observational
data
• Learn methods for prediction
• Focus on applications – theory is used only as needed to
understand the whys of the methods
• Learn to evaluate the regression analysis of others – this means
you will be able to read/understand empirical economic papers in
other economic courses
• Get some hands-on experience with regression analysis in your
problem sets

Copyright © 2019 Pearson Education Ltd. All Rights Reserved.