Introduction to Econometrics

Gary Koop is Professor of Economics at the University of Strathclyde. Gary has published numerous articles econometrics in journals such as the Journal of Econometrics and Journal of Applied Econometrics. Gary has taught econometrics for many years and is the author of following textbooks, all published by John Wiley & Sons Ltd: Analysis of Economic Data 2ed, Analysis of Financial Data and Bayesian Econometrics

“An introductory text offering econometric methodology for quantifying and managing this variety of risk, illustrated by empirical examples.” (Times Higher Education Supplement, Thursday 28th February)

• Preface ixChapter 1 An Overview of Econometrics 11.1 The importance of econometrics 11.2 Types of economic data 21.3 Working with data: graphical methods 61.4 Working with data: descriptive statistics and correlation 111.5 Chapter summary 26Exercises 26Chapter 2 A Non-technical Introduction to Regression 292.1 Introduction 292.2 The simple regression model 302.3 The multiple regression model 422.4 Chapter summary 55Exercises 57Chapter 3 The Econometrics of the Simple Regression Model 593.1 Introduction 593.2 A review of basic concepts in probability in the context of the regression model 603.3 The classical assumptions for the regression model 643.4 Properties of the ordinary least-squares estimator of β 673.5 Deriving a confidence interval for β 753.6 Hypothesis tests about β 773.7 Modifications to statistical procedures when σ2 is unknown 783.8 Chapter summary 81Exercises 82Appendix 1: Proof of the Gauss–Markov theorem 84Appendix 2: Using asymptotic theory in the simple regression model 85Chapter 4 The Econometrics of the Multiple Regression Model 914.1 Introduction 914.2 Basic results for the multiple regression model 924.3 Issues relating to the choice of explanatory variables 964.4 Hypothesis testing in the multiple regression model 1024.5 Choice of functional form in the multiple regression model 1094.6 Chapter summary 115Exercises 116Appendix: Wald and Lagrange multiplier tests 117Chapter 5 The Multiple Regression Model: Freeing Up the Classical Assumptions 1215.1 Introduction 1215.2 Basic theoretical results 1225.3 Heteroskedasticity 1245.4 The regression model with autocorrelated errors 1385.5 The instrumental variables estimator 1495.6 Chapter summary 164Exercises 165Appendix: Asymptotic results for the OLS and instrumental variables estimators 168Chapter 6 Univariate Time Series Analysis 1736.1 Introduction 1736.2 Time series notation 1756.3 Trends in time series variables 1776.4 The autocorrelation function 1796.5 The autoregressive model 1816.6 Defining stationarity 1956.7 Modeling volatility 1976.8 Chapter summary 205Exercises 207Appendix: MA and ARMA models 210Chapter 7 Regression with Time Series Variables 2137.1 Introduction 2137.2 Time series regression when X and Yare stationary 2147.3 Time series regression when Y and X have unit roots 2177.4 Time series regression when Y and X have unit roots but are NOTcointegrated 2277.5 Granger causality 2277.6 Vector autoregressions 2337.7 Chapter summary 247Exercises 248Appendix: The theory of forecasting 251Chapter 8 Models for Panel Data 2558.1 Introduction 2558.2 The pooled model 2568.3 Individual effects models 2568.4 Chapter summary 271Exercises 272Chapter 9 Qualitative Choice and Limited Dependent Variable Models 2779.1 Introduction 2779.2 Qualitative choice models 2789.3 Limited dependent variable models 2969.4 Chapter summary 304Exercises 306Chapter 10 Bayesian Econometrics 30910.1 An overview of Bayesian econometrics 30910.2 The normal linear regression model with natural conjugate prior and a single explanatory variable 31510.3 Chapter summary 326Exercises 326Appendix: Bayesian analysis of the simple regression model with unknown variance 328Appendix A: Mathematical Basics 333Appendix B: Probability Basics 338Appendix C: Basic Concepts in Asymptotic Theory 348Appendix D: Writing an Empirical Project 353Tables 359Table 1. Area under the standard normal distribution Pr(0 ≤ Z ≤ z) 359Table 2. Area under the Student t distribution for different degrees of freedom (DF), Pr(Z ≥ z) = α 360Table 3. Percentiles of the chi-square distribution 361Table 4a. Area under the F-distribution for different degrees of freedom, ν1 and ν2, Pr(Z ≥ z) = 0.05 362Table 4b. Area under the F-distribution for different degrees of freedom, ν1 and ν2, Pr(Z ≥ z) = 0.01 363Bibliography 364Index 365