Analysis of Financial Data teaches basic methods and techniques of data analysis to finance students. It covers many of the major tools used by the financial economist i.e. regression and time series methods including discussion of nonstationary models, multivariate concepts such as cointegration and models of conditional volatility.

Preface ix

Chapter 1 Introduction 1

Organization of the book 3

Useful background 4

Appendix 1.1: Concepts in mathematics used in this book 4

Chapter 2 Basic data handling 9

Types of financial data 9

Obtaining data 15

Working with data: graphical methods 16

Working with data: descriptive statistics 21

Expected values and variances 24

Chapter summary 26

Appendix 2.1: Index numbers 27

Appendix 2.2: Advanced descriptive statistics 30

Chapter 3 Correlation 33

Understanding correlation 33

Understanding why variables are correlated 39

Understanding correlation through XY-plots 40

Correlation between several variables 44

Covariances and population correlations 45

Chapter summary 47

Appendix 3.1: Mathematical details 47

Chapter 4 An introduction to simple regression 49

Regression as a best fitting line 50

Interpreting OLS estimates 53

Fitted values and R²: measuring the fit of a regression model 55

Nonlinearity in regression 61

Chapter summary 64

Appendix 4.1: Mathematical details 65

Chapter 5 Statistical aspects of regression 69

Which factors affect the accuracy of the estimate βˆ? 70

Calculating a confidence interval for β 73

Testing whether β =0 79

Hypothesis testing involving R²: the F-statistic 84

Chapter summary 86

Appendix 5.1: Using statistical tables for testing whether β =0 87

Chapter 6 Multiple regression 91

Regression as a best fitting line 93

Ordinary least squares estimation of the multiple regression model 93

Statistical aspects of multiple regression 94

Interpreting OLS estimates 95

Pitfalls of using simple regression in a multiple regression context 98

Omitted variables bias 100

Multicollinearity 102

Chapter summary 105

Appendix 6.1: Mathematical interpretation of regression coefficients 105

Chapter 7 Regression with dummy variables 109

Simple regression with a dummy variable 112

Multiple regression with dummy variables 114

Multiple regression with both dummy and non-dummy explanatory variables 116

Interacting dummy and non-dummy variables 120

What if the dependent variable is a dummy? 121

Chapter summary 122

Chapter 8 Regression with lagged explanatory variables 123

Aside on lagged variables 125

Aside on notation 127

Selection of lag order 132

Chapter summary 135

Chapter 9 Univariate time series analysis 137

The autocorrelation function 140

The autoregressive model for univariate time series 144

Nonstationary versus stationary time series 146

Extensions of the AR(1) model 149

Testing in the AR( p) with deterministic trend model 152

Chapter summary 158

Appendix 9.1: Mathematical intuition for the AR(1) model 159

Chapter 10 Regression with time series variables 161

Time series regression when X and Y are stationary 162

Time series regression when Y and X have unit roots: spurious regression 167

Time series regression when Y and X have unit roots: cointegration 167

Time series regression when Y and X are cointegrated: the error correction model 174

Time series regression when Y and X have unit roots but are not cointegrated 177

Chapter summary 179

Chapter 11 Regression with time series variables with several equations 183

Granger causality 184

Vector autoregressions 190

Chapter summary 203

Appendix 11.1: Hypothesis tests involving more than one coefficient 204

Appendix 11.2: Variance decompositions 207

Chapter 12 Financial volatility 211

Volatility in asset prices: Introduction 212

Autoregressive conditional heteroskedasticity (ARCH) 217

Chapter summary 222

Appendix A Writing an empirical project 223

Description of a typical empirical project 223

General considerations 225

Appendix B Data directory 227

Index 231

Analysis of Financial Data teaches the basic methods and techniques of data analysis to finance students, by showing them how to apply such techniques in the context of real-world empirical problems.

Adopting a largely non-mathematical approach Analysis of Financial Data relies more on verbal intuition and graphical methods for understanding.

Key features include:

• Coverage of many of the major tools used by the financial economist e.g. correlation, regression, time series analysis and methods for analyzing financial volatility.
• Extensive use of real data examples, which involves readers in hands-on computer work.
• Mathematical techniques at a level suited to MBA students and undergraduates taking a first course in the topic.

Supplementary material for readers and lecturers provided on an accompanying website.

Preface. Chapter 1: Introduction. Organization of the book. Useful background. Appendix 1.1: Concepts in mathematics used in this book. Chapter 2: Basic data handling. Types of financial data. Obtaining data. Working with data: graphical methods. Working with data: descriptive statistics. Expected values and variances. Chapter summary. Appendix 2.1: Index numbers. Appendix 2.2: Advanced descriptive statistics. Chapter 3: Correlation. Understanding correlation. Understanding why variables are correlated. Understanding correlation through XY-plots. Correlation between several variables. Covariances and population correlations. Chapter summary. Appendix 3.1: Mathematical details. Chapter 4: An introduction to simple regression. Regression as a best fitting line. Interpreting OLS estimates. Fitted values and R2: measuring the fit of a regression model. Nonlinearity in regression. Chapter summary. Appendix 4.1: Mathematical details. Chapter 5: Statistical aspects of regression. Which factors affect the accuracy of the estimate β? Calculating a confidence interval for β. Testing whether β = 0. Hypothesis testing involving R2: the F-statistic. Chapter summary. Appendix 5.1: Using statistical tables for testing whether β = 0. Chapter 6: Multiple regression. Regression as a best fitting line. Ordinary least squares estimation of the multiple regression model. Statistical aspects of multiple regression. Interpreting OLS estimates. Pitfalls of using simple regression in a multiple regression context. Omitted variables bias. Multicollinearity. Chapter summary. Appendix 6.1: Mathematical interpretation of regression coefficients. Chapter 7: Regression with dummy variables. Simple regression with a dummy variable. Multiple regression with dummy variables. Multiple regression with both dummy and non-dummy explanatory variables. Interacting dummy and non-dummy variables. What if the dependent variable is a dummy? Chapter summary. Chapter 8: Regression with lagged explanatory variables. Aside on lagged variables. Aside on notation. Selection of lag order. Chapter summary. Chapter 9: Univariate time series analysis. The autocorrelation function. The autoregressive model for univariate time series. Nonstationary versus stationary time series. Extensions of the AR(1) model. Testing in the AR( p) with deterministic trend model. Chapter summary. Appendix 9.1: Mathematical intuition for the AR(1) model. Chapter 10: Regression with time series variables. Time series regression when X and Y are stationary. Time series regression when Y and X have unit roots: spurious regression. Time series regression when Y and X have unit roots: cointegration. Time series regression when Y and X are cointegrated: the error correction model. Time series regression when Y and X have unit roots but are not cointegrated. Chapter summary. Chapter 11: Regression with time series variables with several equations. Granger causality. Vector autoregressions. Chapter summary. Appendix 11.1: Hypothesis tests involving more than one coefficient. Appendix 11.2: Variance decompositions. Chapter 12: Financial volatility. Volatility in asset prices: Introduction. Autoregressive conditional heteroskedasticity (ARCH). Chapter summary. Appendix A: Writing an empirical project. Description of a typical empirical project. General considerations. Appendix B: Data directory. Index.