Preface
1. Modeling Social Dynamics
1.1 Time Series Analysis
1.1.1 Contending Approaches and the Quasi-Experimental Tradition
1.1.2 The Architecture of the Book
1.2 Challenges in Time Series Modeling
1.2.1 Time Series Measurement
1.2.2 Fit: On the Use of Time Trends and Counters
1.2.3 Discerning Structural Change
1.3 The First Step
1.4 For Further Reading
2. Univariate Time Series Models
2.1 Understanding Univariate Processes
2.1.1 Why Stochastic Models of Time Series?
2.1.2 White Noise
2.1.3 A Note on Notation
2.2 Univariate Processes
2.2.1 The Autoregressive (AR) Process
2.2.2 The Moving Average (MA) Process
2.2.3 The Equivalence of AR and MA Processes
2.3 Diagnosing Stationary Univariate Processes
2.3.1 The Autocorrelation Function (ACF)
2.3.2 The Partial Autocorrelation Function (PACF)
2.3.3 Seasonality
2.4 Model Estimation, Interpretation, and Diagnosis
2.4.1 Residual Analysis
2.4.2 Metadiagnosis
2.5 Application: U.S. Homicide Rates
2.6 Application: U.S. Relations with the Middle East
2.7 Intervention Models
2.7.1 Impact Assessment: Continuing the Middle East Example
2.7.2 Structural Breaks
2.8 Conclusion
2.9 For Further Reading
3. Dynamic Regression Models
3.1 Variations of the General Dynamic Regression Model
3.1.1 Building Blocks
3.1.2 Autoregressive Models/Lagged Endogenous Variables
3.1.3 The Workhorse Model
3.1.4 Serial Correlation: Tests and Solutions
3.2 Obtaining Consistent Estimates
3.2.1 Approach Number One: Pseudo-GLS Estimation
3.2.2 Approach Number Two: Hatanaka's Method
3.3 Illustration: Explaining Voting Intentions in the United Kingdom
3.3.1 Voting Intentions and Exchange Rates: Results
3.4 Dynamic Regression Models with Multiple Causal Variable Lags
3.4.1 Distributed Lag Models
3.4.2 Autoregressive Distributed Lag Models
3.5 Conclusion
3.6 For Further Reading
4. Modeling the Dynamics of Social Systems
4.1 Two Approaches to Multivariate Time Series Analysis
4.2 The Structural Equation Approach
4.2.1 The Workhorse Model of Time Series Regression Revisited
4.2.2 The Structural Equation Approach in Practice
4.2.3 Application: The Political Economy of Monetary Policy in
the United Kingdom
4.3 Vector Autoregression
4.3.1 Terminology and Stationarity
4.3.2 Specification and Estimation
4.3.3 Tools for Inference in VAR Modeling
4.3.4 VAR Modeling in Practice
4.4 The Political Economy of Monetary Policy in the United Kingdom
Revisited
4.5 Conclusion: SEQ and VAR Modeling Compared
4.6 For Further Reading
5. Univariate, Nonstationary Processes: Tests and Modeling
5.1 Stationary Data
5.2 Stationarity and Unit Root Tests
5.2.1 Dickey-Fuller and Augmented Dickey-Fuller Tests
5.2.2 Variance Ratio Test
5.2.3 Modified Rescaled Range Test
5.2.4 KPSS Test
5.2.5 Double Unit Roots
5.3 Application: Macropartisanship
5.3.1 Macropartisanship: Diagnostic Plots
5.3.2 Macropartisanship: Formal Statistical Tests
5.3.3 Macropartisanship: An ARIMA Model
5.4 Conclusion
5.5 For Further Reading
6. Cointegration and Error Correction Models
6.1 Introduction
6.2 The Intuition behind Cointegration
6.3 Estimating an ECM
6.3.1 Engle-Granger Regression-Based Approach
6.3.2 Johansen's VAR Approach
6.3.3 Application: The Engle-Granger Methodology and the Indo-Pakistani Arms Race
6.4 ECMs for Non-Integrated Data?
6.5 Conclusion
6.6 For Further Reading
7. Selections on Time Series Analysis
7.1 Fractional Integration
7.1.1 Fractional Cointegration
7.1.2 Near and Fractional Integration
7.2 Incorporating Heterogeneity
7.3 Forecasting
7.3.1 Purpose and Definitions
7.3.2 Approaches to Forecasting
7.3.3 Criteria for Determining Optimum Forecast
7.3.4 Illustration: Forecasts of the U.S. Uniform Crime Reports
Homicide Data, 1975-1993
7.3.5 Forecasting Discussion
7.4 Estimating and Modeling with Unknown Structural Breaks
7.4.1 Identifying a Single Unknown Structural Break
7.4.2 Identifying Multiple Unknown Structural Breaks
7.4.3 Hypothesis Tests for Multiple Structural Breaks
7.5 Conclusion
7.6 For Further Reading
8. Concluding Thoughts for the Time Series Analyst
8.1 A Review
8.2 Looking Ahead
Appendix: Time Series Models as Difference Equations
A.1 Introduction
A.2 The Nature of Difference Equation Models
A.2.1 Expressing Difference Equations
A.2.2 The Solutions of Difference Equations
A.3 Finding the Solutions of Difference Equation Models
A.3.1 Solving the Most Simple First-Order Difference Equations
A.3.2 Solving More Complicated Difference Equations
A.3.3 Solving Systems of Difference Equations
A.4 Bring on the Data