Analysis of Financial Data

Gary Koop is Professor of Economics at the University of Strathclyde.

• Preface ixChapter 1 Introduction 1Organization of the book 3Useful background 4Appendix 1.1: Concepts in mathematics used in this book 4Chapter 2 Basic data handling 9Types of financial data 9Obtaining data 15Working with data: graphical methods 16Working with data: descriptive statistics 21Expected values and variances 24Chapter summary 26Appendix 2.1: Index numbers 27Appendix 2.2: Advanced descriptive statistics 30Chapter 3 Correlation 33Understanding correlation 33Understanding why variables are correlated 39Understanding correlation through XY-plots 40Correlation between several variables 44Covariances and population correlations 45Chapter summary 47Appendix 3.1: Mathematical details 47Chapter 4 An introduction to simple regression 49Regression as a best fitting line 50Interpreting OLS estimates 53Fitted values and R2: measuring the fit of a regression model 55Nonlinearity in regression 61Chapter summary 64Appendix 4.1: Mathematical details 65Chapter 5 Statistical aspects of regression 69Which factors affect the accuracy of the estimate βˆ? 70Calculating a confidence interval for β 73Testing whether β =0 79Hypothesis testing involving R2: the F-statistic 84Chapter summary 86Appendix 5.1: Using statistical tables for testing whether β =0 87Chapter 6 Multiple regression 91Regression as a best fitting line 93Ordinary least squares estimation of the multiple regression model 93Statistical aspects of multiple regression 94Interpreting OLS estimates 95Pitfalls of using simple regression in a multiple regression context 98Omitted variables bias 100Multicollinearity 102Chapter summary 105Appendix 6.1: Mathematical interpretation of regression coefficients 105Chapter 7 Regression with dummy variables 109Simple regression with a dummy variable 112Multiple regression with dummy variables 114Multiple regression with both dummy and non-dummy explanatory variables 116Interacting dummy and non-dummy variables 120What if the dependent variable is a dummy? 121Chapter summary 122Chapter 8 Regression with lagged explanatory variables 123Aside on lagged variables 125Aside on notation 127Selection of lag order 132Chapter summary 135Chapter 9 Univariate time series analysis 137The autocorrelation function 140The autoregressive model for univariate time series 144Nonstationary versus stationary time series 146Extensions of the AR(1) model 149Testing in the AR( p) with deterministic trend model 152Chapter summary 158Appendix 9.1: Mathematical intuition for the AR(1) model 159Chapter 10 Regression with time series variables 161Time series regression when X and Y are stationary 162Time series regression when Y and X have unit roots: spurious regression 167Time series regression when Y and X have unit roots: cointegration 167Time series regression when Y and X are cointegrated: the error correction model 174Time series regression when Y and X have unit roots but are not cointegrated 177Chapter summary 179Chapter 11 Regression with time series variables with several equations 183Granger causality 184Vector autoregressions 190Chapter summary 203Appendix 11.1: Hypothesis tests involving more than one coefficient 204Appendix 11.2: Variance decompositions 207Chapter 12 Financial volatility 211Volatility in asset prices: Introduction 212Autoregressive conditional heteroskedasticity (ARCH) 217Chapter summary 222Appendix A Writing an empirical project 223Description of a typical empirical project 223General considerations 225Appendix B Data directory 227Index 231