Applied Econometrics

Dimitrios Asteriou, Stephen G. Hall

• Publisher: Bloomsbury Academic
• Year: 2021
• Print ISBN: 9781352012026
• eBook ISBN: 9781350306141
• Edition: 4
• Topic: Economics
• Subtopic: Econometrics
• Index: Economics

This trusted textbook returns in its 4th edition with even more exercises to help consolidate understanding - and a companion website featuring additional materials, including a solutions manual for instructors. Offering a unique blend of theory and practical application, it provides ideal preparation for doing applied econometric work as it takes students from a basic level up to an advanced understanding in an intuitive, step-by-step fashion. Clear presentation of economic tests and methods of estimation is paired with practical guidance on using several types of software packages. Using real world data throughout, the authors place emphasis upon the interpretation of results, and the conclusions to be drawn from them in econometric work. This book will be essential reading for economics undergraduate and master's students taking a course in applied econometrics. Its practical nature makes it ideal for modules requiring a research project. New to this Edition:
- Additional practical exercises throughout to help consolidate understanding
- A freshly-updated companion website featuring a new solutions manual for instructors

Part I

Statistical Background and Basic Data Handling

1 Fundamental Concepts

2 The Structure of Economic Data and Basic Data Handling

1	Fundamental Concepts

CHAPTER CONTENTS

Introduction

A simple example

A statistical framework

Properties of the sampling distribution of the mean

Hypothesis testing and the central limit theorem

Conclusion

Introduction

This chapter outlines some of the fundamental concepts that lie behind much of the rest of this book, including the ideas of a population distribution and a sampling distribution, the importance of random sampling, the law of large numbers and the central limit theorem. It then goes on to show how these ideas underpin the conventional approach to testing hypotheses and constructing confidence intervals.

Econometrics has a number of roles in terms of forecasting and analysing real data and problems. At the core of these roles, however, is the desire to pin down the magnitudes of effects and test their significance. Economic theory often points to the direction of a causal relationship (if income rises we may expect consumption to rise), but theory rarely suggests an exact magnitude. Yet, in a policy or business context, having a clear idea of the magnitude of an effect may be extremely important, and this is the realm of econometrics.

The aim of this chapter is to clarify some basic definitions and ideas in order to give the student an intuitive understanding of these underlying concepts. The account given here will therefore deliberately be less formal than much of the material later in the book.

A simple example

Consider a very simple example to illustrate the idea we are putting forward here. Table 1.1 shows the average age at death for both men and women in the 15 European countries that made up the European Union (EU) before its enlargement.

Simply looking at these figures makes it fairly obvious that women can expect to live longer than men in each of these countries, and if we take the average across all countries we can clearly see that again, on a Europe-wide basis, women tend to live longer than men. However, there is quite considerable variation between the countries, and it might be reasonable to ask whether in general, in the world population, we would expect women to live longer than men.

Table 1.1 Average age at death for the EU15 countries (2002)

	Women	Men
Austria	81.2	75.4
Belgium	81.4	75.1
Denmark	79.2	74.5
Finland	81.5	74.6
France	83.0	75.5
Germany	80.8	74.8
Greece	80.7	75.4
Ireland	78.5	73.0
Italy	82.9	76.7
Luxembourg	81.3	74.9
Netherlands	80.6	75.5
Portugal	79.4	72.4
Spain	82.9	75.6
Sweden	82.1	77.5
UK	79.7	75.0
beginlwmath26pt] Mean	81.0	75.1
Standard deviation	1.3886616	1.2391241

A natural way to approach this would be to look at the difference in the mean life expectancy between men and women for the whole of Europe and to ask whether this is significantly different from zero. This involves a number of fundamental steps: first the difference in average life expectancy has to be estimated, then a measure of its uncertainty must be constructed, and finally the hypothesis that the difference is zero needs to be tested.

Table 1.1 gives the average (or mean) life expectancy for men and women for the EU as a whole, simply defined as:

$\begin{array}{ll}{\overline{Y}}_w=\frac{1}{15}{\displaystyle \sum _{i=1}^{15}{Y}_{wi}}\hfill & {\overline{Y}}_m=\frac{1}{15}{\displaystyle \sum _{i=1}^{15}{Y}_{mi}}\hfill \end{array}$ (1.1)

where Ȳ_w is the EU average life expectancy for women and Ȳ_m is the EU average life expectancy for men. A natural estimate of the difference between the two means is (Ȳ_w − Ȳ_m). Table 1.1 also gives the average dispersion for each of these means, defined as the standard deviation, which is given by:

$\begin{array}{ll}S.D{.}_j=\sqrt{{\displaystyle \sum _{i=1}^{15}{\left(Y_{ji}-{\overline{Y}}_j\right)}^2}}\hfill & j=w,m\hfill \end{array}$ (1.2)

As we have an estimate of the difference and an estimate of the uncertainty of our measures, we can now construct a formal hypothesis test. The test for the difference between two means is:

...

Table of contents

• Cover
• Titlepage
• Dedication
• Brief Contents
• Table of Contents
• List of Figures
• List of Tables
• Preface
• Acknowledgements
• Part I Statistical Background and Basic Data Handling
• Part II The Classical Linear Regression Model
• Part III Violating the Assumptions of the CLRM
• Part IV Topics in Econometrics
• Part V Time Series Econometrics
• Part VI Panel Data Econometrics
• Part VII Using Econometric Software
• Appendix: Statistical Tables
• Bibliography
• Index
• eCopyright