RUEY S. TSAY, PhD, is H. G. B. Alexander Professor of Econometrics and Statistics at the University of Chicago Booth School of Business. Dr. Tsay has written over 100 published articles in the areas of business and economic forecasting, data analysis, risk management, and process control, and he is the coauthor of A Course in Time Series Analysis (Wiley). Dr. Tsay is a Fellow of the American Statistical Association, the Institute of Mathematical Statistics, the Royal Statistical Society, and Academia Sinica.

• Preface xviiPreface to the Second Edition xixPreface to the First Edition xxi1 Financial Time Series and Their Characteristics 11.1 Asset Returns, 21.2 Distributional Properties of Returns, 71.3 Processes Considered, 222 Linear Time Series Analysis and Its Applications 292.1 Stationarity, 302.2 Correlation and Autocorrelation Function, 302.3 White Noise and Linear Time Series, 362.4 Simple AR Models, 372.5 Simple MA Models, 572.6 Simple ARMA Models, 642.7 Unit-Root Nonstationarity, 712.8 Seasonal Models, 812.9 Regression Models with Time Series Errors, 902.10 Consistent Covariance Matrix Estimation, 972.11 Long-Memory Models, 1013 Conditional Heteroscedastic Models 1093.1 Characteristics of Volatility, 1103.2 Structure of a Model, 1113.3 Model Building, 1133.4 The ARCH Model, 1153.5 The GARCH Model, 1313.6 The Integrated GARCH Model, 1403.7 The GARCH-M Model, 1423.8 The Exponential GARCH Model, 1433.9 The Threshold GARCH Model, 1493.10 The CHARMA Model, 1503.11 Random Coefficient Autoregressive Models, 1523.12 Stochastic Volatility Model, 1533.13 Long-Memory Stochastic Volatility Model, 1543.14 Application, 1553.15 Alternative Approaches, 1593.16 Kurtosis of GARCH Models, 1654 Nonlinear Models and Their Applications 1754.1 Nonlinear Models, 1774.2 Nonlinearity Tests, 2054.3 Modeling, 2144.4 Forecasting, 2154.5 Application, 2185 High-Frequency Data Analysis and Market Microstructure 2315.1 Nonsynchronous Trading, 2325.2 Bid–Ask Spread, 2355.3 Empirical Characteristics of Transactions Data, 2375.4 Models for Price Changes, 2445.5 Duration Models, 2535.6 Nonlinear Duration Models, 2645.7 Bivariate Models for Price Change and Duration, 2655.8 Application, 2706 Continuous-Time Models and Their Applications 2876.1 Options, 2886.2 Some Continuous-Time Stochastic Processes, 2886.3 Ito's Lemma, 2926.4 Distributions of Stock Prices and Log Returns, 2976.5 Derivation of Black–Scholes Differential Equation, 2986.6 Black–Scholes Pricing Formulas, 3006.7 Extension of Ito's Lemma, 3096.8 Stochastic Integral, 3106.9 Jump Diffusion Models, 3116.10 Estimation of Continuous-Time Models, 3187 Extreme Values, Quantiles, and Value at Risk 3257.1 Value at Risk, 3267.2 RiskMetrics, 3287.3 Econometric Approach to VaR Calculation, 3337.4 Quantile Estimation, 3387.5 Extreme Value Theory, 3427.6 Extreme Value Approach to VaR, 3537.7 New Approach Based on the Extreme Value Theory, 3597.8 The Extremal Index, 3778 Multivariate Time Series Analysis and Its Applications 3898.1 Weak Stationarity and Cross-Correlation Matrices, 3908.2 Vector Autoregressive Models, 3998.3 Vector Moving-Average Models, 4178.4 Vector ARMA Models, 4228.5 Unit-Root Nonstationarity and Cointegration, 4288.6 Cointegrated VAR Models, 4328.7 Threshold Cointegration and Arbitrage, 4428.8 Pairs Trading, 4469 Principal Component Analysis and Factor Models 4679.1 A Factor Model, 4689.2 Macroeconometric Factor Models, 4709.3 Fundamental Factor Models, 4769.4 Principal Component Analysis, 4839.5 Statistical Factor Analysis, 4899.6 Asymptotic Principal Component Analysis, 49810 Multivariate Volatility Models and Their Applications 50510.1 Exponentially Weighted Estimate, 50610.2 Some Multivariate GARCH Models, 51010.3 Reparameterization, 51610.4 GARCH Models for Bivariate Returns, 52110.5 Higher Dimensional Volatility Models, 53710.6 Factor–Volatility Models, 54310.7 Application, 54610.8 Multivariate t Distribution, 54811 State-Space Models and Kalman Filter 55711.1 Local Trend Model, 55811.2 Linear State-Space Models, 57611.3 Model Transformation, 57711.4 Kalman Filter and Smoothing, 59111.5 Missing Values, 60011.6 Forecasting, 60111.7 Application, 60212 Markov Chain Monte Carlo Methods with Applications 61312.1 Markov Chain Simulation, 61412.2 Gibbs Sampling, 61512.3 Bayesian Inference, 61712.4 Alternative Algorithms, 62212.5 Linear Regression with Time Series Errors, 62412.6 Missing Values and Outliers, 62812.7 Stochastic Volatility Models, 63612.8 New Approach to SV Estimation, 64912.9 Markov Switching Models, 66012.10 Forecasting, 66612.11 Other Applications, 669Exercises, 670References, 671Index 673

"Analysis of financial time series, third edition, is an ideal book for introductory courses on time series at the graduate level and a valuable supplement for statistics courses in time series at the upper-undergraduate level." (Mathematical Reviews, 2011) "Nevertheless, all in all the book can be a very useful reference for students as well as for professionals." (Zentralblatt MATH, 2011)"Factor models, an important technique used in quantitative finance, are given a full treatment with macroeconomic factor models and fundamental factor models. The coverage of the book is comprehensive. It starts from basic time series techniques and finishes with advanced concepts such as state space models and MCMC methods. There is a balance between the theoretical background necessary to appreciate the nuances and the practical aspect of implementation. More importantly it gives insights about what time series models can't address. The book has an excellent supporting website which has all the programs and data sets which helps to internalize the concepts. Finally, teaching professionals should find the solutions manual as a valuable tool to explain concepts and to ensure understanding." (BookPleasures.com, January 2011)"This book provides a broad, mature, and systematic introduction to current financial econometric models and their applications to modeling and prediction of financial time series data. It utilizes real-world examples and real financial data throughout the book to apply the models and methods described." (Insurance News Net, 8 December 2010)

RUEY S. TSAY, PhD, is H. G. B. Alexander Professor of Econometrics and Statistics at the University of Chicago Booth School of Business. Dr. Tsay has written over 100 published articles in the areas of business and economic forecasting, data analysis, risk management, and process control, and he is the coauthor of A Course in Time Series Analysis (Wiley). Dr. Tsay is a Fellow of the American Statistical Association, the Institute of Mathematical Statistics, the Royal Statistical Society, and Academia Sinica.

• Preface xviiPreface to the Second Edition xixPreface to the First Edition xxi1 Financial Time Series and Their Characteristics 11.1 Asset Returns, 21.2 Distributional Properties of Returns, 71.3 Processes Considered, 222 Linear Time Series Analysis and Its Applications 292.1 Stationarity, 302.2 Correlation and Autocorrelation Function, 302.3 White Noise and Linear Time Series, 362.4 Simple AR Models, 372.5 Simple MA Models, 572.6 Simple ARMA Models, 642.7 Unit-Root Nonstationarity, 712.8 Seasonal Models, 812.9 Regression Models with Time Series Errors, 902.10 Consistent Covariance Matrix Estimation, 972.11 Long-Memory Models, 1013 Conditional Heteroscedastic Models 1093.1 Characteristics of Volatility, 1103.2 Structure of a Model, 1113.3 Model Building, 1133.4 The ARCH Model, 1153.5 The GARCH Model, 1313.6 The Integrated GARCH Model, 1403.7 The GARCH-M Model, 1423.8 The Exponential GARCH Model, 1433.9 The Threshold GARCH Model, 1493.10 The CHARMA Model, 1503.11 Random Coefficient Autoregressive Models, 1523.12 Stochastic Volatility Model, 1533.13 Long-Memory Stochastic Volatility Model, 1543.14 Application, 1553.15 Alternative Approaches, 1593.16 Kurtosis of GARCH Models, 1654 Nonlinear Models and Their Applications 1754.1 Nonlinear Models, 1774.2 Nonlinearity Tests, 2054.3 Modeling, 2144.4 Forecasting, 2154.5 Application, 2185 High-Frequency Data Analysis and Market Microstructure 2315.1 Nonsynchronous Trading, 2325.2 Bid–Ask Spread, 2355.3 Empirical Characteristics of Transactions Data, 2375.4 Models for Price Changes, 2445.5 Duration Models, 2535.6 Nonlinear Duration Models, 2645.7 Bivariate Models for Price Change and Duration, 2655.8 Application, 2706 Continuous-Time Models and Their Applications 2876.1 Options, 2886.2 Some Continuous-Time Stochastic Processes, 2886.3 Ito's Lemma, 2926.4 Distributions of Stock Prices and Log Returns, 2976.5 Derivation of Black–Scholes Differential Equation, 2986.6 Black–Scholes Pricing Formulas, 3006.7 Extension of Ito's Lemma, 3096.8 Stochastic Integral, 3106.9 Jump Diffusion Models, 3116.10 Estimation of Continuous-Time Models, 3187 Extreme Values, Quantiles, and Value at Risk 3257.1 Value at Risk, 3267.2 RiskMetrics, 3287.3 Econometric Approach to VaR Calculation, 3337.4 Quantile Estimation, 3387.5 Extreme Value Theory, 3427.6 Extreme Value Approach to VaR, 3537.7 New Approach Based on the Extreme Value Theory, 3597.8 The Extremal Index, 3778 Multivariate Time Series Analysis and Its Applications 3898.1 Weak Stationarity and Cross-Correlation Matrices, 3908.2 Vector Autoregressive Models, 3998.3 Vector Moving-Average Models, 4178.4 Vector ARMA Models, 4228.5 Unit-Root Nonstationarity and Cointegration, 4288.6 Cointegrated VAR Models, 4328.7 Threshold Cointegration and Arbitrage, 4428.8 Pairs Trading, 4469 Principal Component Analysis and Factor Models 4679.1 A Factor Model, 4689.2 Macroeconometric Factor Models, 4709.3 Fundamental Factor Models, 4769.4 Principal Component Analysis, 4839.5 Statistical Factor Analysis, 4899.6 Asymptotic Principal Component Analysis, 49810 Multivariate Volatility Models and Their Applications 50510.1 Exponentially Weighted Estimate, 50610.2 Some Multivariate GARCH Models, 51010.3 Reparameterization, 51610.4 GARCH Models for Bivariate Returns, 52110.5 Higher Dimensional Volatility Models, 53710.6 Factor–Volatility Models, 54310.7 Application, 54610.8 Multivariate t Distribution, 54811 State-Space Models and Kalman Filter 55711.1 Local Trend Model, 55811.2 Linear State-Space Models, 57611.3 Model Transformation, 57711.4 Kalman Filter and Smoothing, 59111.5 Missing Values, 60011.6 Forecasting, 60111.7 Application, 60212 Markov Chain Monte Carlo Methods with Applications 61312.1 Markov Chain Simulation, 61412.2 Gibbs Sampling, 61512.3 Bayesian Inference, 61712.4 Alternative Algorithms, 62212.5 Linear Regression with Time Series Errors, 62412.6 Missing Values and Outliers, 62812.7 Stochastic Volatility Models, 63612.8 New Approach to SV Estimation, 64912.9 Markov Switching Models, 66012.10 Forecasting, 66612.11 Other Applications, 669Exercises, 670References, 671Index 673

"Analysis of financial time series, third edition, is an ideal book for introductory courses on time series at the graduate level and a valuable supplement for statistics courses in time series at the upper-undergraduate level." (Mathematical Reviews, 2011) "Nevertheless, all in all the book can be a very useful reference for students as well as for professionals." (Zentralblatt MATH, 2011)"Factor models, an important technique used in quantitative finance, are given a full treatment with macroeconomic factor models and fundamental factor models. The coverage of the book is comprehensive. It starts from basic time series techniques and finishes with advanced concepts such as state space models and MCMC methods. There is a balance between the theoretical background necessary to appreciate the nuances and the practical aspect of implementation. More importantly it gives insights about what time series models can't address. The book has an excellent supporting website which has all the programs and data sets which helps to internalize the concepts. Finally, teaching professionals should find the solutions manual as a valuable tool to explain concepts and to ensure understanding." (BookPleasures.com, January 2011)"This book provides a broad, mature, and systematic introduction to current financial econometric models and their applications to modeling and prediction of financial time series data. It utilizes real-world examples and real financial data throughout the book to apply the models and methods described." (Insurance News Net, 8 December 2010)

RUEY S. TSAY, PhD, is H. G. B. Alexander Professor of Econometrics and Statistics at the University of Chicago Booth School of Business. Dr. Tsay has written over 100 published articles in the areas of business and economic forecasting, data analysis, risk management, and process control, and he is the coauthor of A Course in Time Series Analysis (Wiley). Dr. Tsay is a Fellow of the American Statistical Association, the Institute of Mathematical Statistics, the Royal Statistical Society, and Academia Sinica.

• Preface xviiPreface to the Second Edition xixPreface to the First Edition xxi1 Financial Time Series and Their Characteristics 11.1 Asset Returns, 21.2 Distributional Properties of Returns, 71.3 Processes Considered, 222 Linear Time Series Analysis and Its Applications 292.1 Stationarity, 302.2 Correlation and Autocorrelation Function, 302.3 White Noise and Linear Time Series, 362.4 Simple AR Models, 372.5 Simple MA Models, 572.6 Simple ARMA Models, 642.7 Unit-Root Nonstationarity, 712.8 Seasonal Models, 812.9 Regression Models with Time Series Errors, 902.10 Consistent Covariance Matrix Estimation, 972.11 Long-Memory Models, 1013 Conditional Heteroscedastic Models 1093.1 Characteristics of Volatility, 1103.2 Structure of a Model, 1113.3 Model Building, 1133.4 The ARCH Model, 1153.5 The GARCH Model, 1313.6 The Integrated GARCH Model, 1403.7 The GARCH-M Model, 1423.8 The Exponential GARCH Model, 1433.9 The Threshold GARCH Model, 1493.10 The CHARMA Model, 1503.11 Random Coefficient Autoregressive Models, 1523.12 Stochastic Volatility Model, 1533.13 Long-Memory Stochastic Volatility Model, 1543.14 Application, 1553.15 Alternative Approaches, 1593.16 Kurtosis of GARCH Models, 1654 Nonlinear Models and Their Applications 1754.1 Nonlinear Models, 1774.2 Nonlinearity Tests, 2054.3 Modeling, 2144.4 Forecasting, 2154.5 Application, 2185 High-Frequency Data Analysis and Market Microstructure 2315.1 Nonsynchronous Trading, 2325.2 Bid–Ask Spread, 2355.3 Empirical Characteristics of Transactions Data, 2375.4 Models for Price Changes, 2445.5 Duration Models, 2535.6 Nonlinear Duration Models, 2645.7 Bivariate Models for Price Change and Duration, 2655.8 Application, 2706 Continuous-Time Models and Their Applications 2876.1 Options, 2886.2 Some Continuous-Time Stochastic Processes, 2886.3 Ito's Lemma, 2926.4 Distributions of Stock Prices and Log Returns, 2976.5 Derivation of Black–Scholes Differential Equation, 2986.6 Black–Scholes Pricing Formulas, 3006.7 Extension of Ito's Lemma, 3096.8 Stochastic Integral, 3106.9 Jump Diffusion Models, 3116.10 Estimation of Continuous-Time Models, 3187 Extreme Values, Quantiles, and Value at Risk 3257.1 Value at Risk, 3267.2 RiskMetrics, 3287.3 Econometric Approach to VaR Calculation, 3337.4 Quantile Estimation, 3387.5 Extreme Value Theory, 3427.6 Extreme Value Approach to VaR, 3537.7 New Approach Based on the Extreme Value Theory, 3597.8 The Extremal Index, 3778 Multivariate Time Series Analysis and Its Applications 3898.1 Weak Stationarity and Cross-Correlation Matrices, 3908.2 Vector Autoregressive Models, 3998.3 Vector Moving-Average Models, 4178.4 Vector ARMA Models, 4228.5 Unit-Root Nonstationarity and Cointegration, 4288.6 Cointegrated VAR Models, 4328.7 Threshold Cointegration and Arbitrage, 4428.8 Pairs Trading, 4469 Principal Component Analysis and Factor Models 4679.1 A Factor Model, 4689.2 Macroeconometric Factor Models, 4709.3 Fundamental Factor Models, 4769.4 Principal Component Analysis, 4839.5 Statistical Factor Analysis, 4899.6 Asymptotic Principal Component Analysis, 49810 Multivariate Volatility Models and Their Applications 50510.1 Exponentially Weighted Estimate, 50610.2 Some Multivariate GARCH Models, 51010.3 Reparameterization, 51610.4 GARCH Models for Bivariate Returns, 52110.5 Higher Dimensional Volatility Models, 53710.6 Factor–Volatility Models, 54310.7 Application, 54610.8 Multivariate t Distribution, 54811 State-Space Models and Kalman Filter 55711.1 Local Trend Model, 55811.2 Linear State-Space Models, 57611.3 Model Transformation, 57711.4 Kalman Filter and Smoothing, 59111.5 Missing Values, 60011.6 Forecasting, 60111.7 Application, 60212 Markov Chain Monte Carlo Methods with Applications 61312.1 Markov Chain Simulation, 61412.2 Gibbs Sampling, 61512.3 Bayesian Inference, 61712.4 Alternative Algorithms, 62212.5 Linear Regression with Time Series Errors, 62412.6 Missing Values and Outliers, 62812.7 Stochastic Volatility Models, 63612.8 New Approach to SV Estimation, 64912.9 Markov Switching Models, 66012.10 Forecasting, 66612.11 Other Applications, 669Exercises, 670References, 671Index 673

"Analysis of financial time series, third edition, is an ideal book for introductory courses on time series at the graduate level and a valuable supplement for statistics courses in time series at the upper-undergraduate level." (Mathematical Reviews, 2011) "Nevertheless, all in all the book can be a very useful reference for students as well as for professionals." (Zentralblatt MATH, 2011)"Factor models, an important technique used in quantitative finance, are given a full treatment with macroeconomic factor models and fundamental factor models. The coverage of the book is comprehensive. It starts from basic time series techniques and finishes with advanced concepts such as state space models and MCMC methods. There is a balance between the theoretical background necessary to appreciate the nuances and the practical aspect of implementation. More importantly it gives insights about what time series models can't address. The book has an excellent supporting website which has all the programs and data sets which helps to internalize the concepts. Finally, teaching professionals should find the solutions manual as a valuable tool to explain concepts and to ensure understanding." (BookPleasures.com, January 2011)"This book provides a broad, mature, and systematic introduction to current financial econometric models and their applications to modeling and prediction of financial time series data. It utilizes real-world examples and real financial data throughout the book to apply the models and methods described." (Insurance News Net, 8 December 2010)