Cover -- Half-title -- Title -- Copyright -- Contents -- Preface -- Intended Audience and Level -- Organization -- Statistical Software -- References Codes -- Approach -- Acknowledgments -- 1 Introduction -- 1.1 What Are Longitudinal and Panel Data? -- Statistical Modeling -- Defining Longitudinal and Panel Data -- Some Notation -- Prevalence of Longitudinal and Panel Data Analysis -- 1.2 Benefits and Drawbacks of Longitudinal Data -- Dynamic Relationships -- Historical Approach -- Dynamic Relationships and Time-Series Analysis -- Longitudinal Data as Repeated Time Series -- Longitudinal Data as Repeated Cross-Sectional Studies -- Heterogeneity -- Heterogeneity Bias -- Omitted Variables -- Efficiency of Estimators -- Correlation and Causation -- Drawbacks: Attrition -- 1.3 Longitudinal Data Models -- Types of Inference -- Social Science Statistical Modeling -- Modeling Issues -- Types of Applications -- 1.4 Historical Notes -- 2 Fixed-Effects Models -- 2.1 Basic Fixed-Effects Model -- Data -- Basic Models -- Parameters of Interest -- Subject and Time Heterogeneity -- 2.2 Exploring Longitudinal Data -- Why Explore? -- Data Exploration Techniques -- Multiple Time-Series Plots -- Scatter Plots with Symbols -- Basic Added-Variable Plot -- Trellis Plot -- 2.3 Estimation and Inference -- Least-Squares Estimation -- Other Properties of Estimators -- ANOVA Table and Standard Errors -- Large-Sample Properties of Estimators -- 2.4 Model Specification and Diagnostics -- 2.4.1 Pooling Test -- 2.4.2 Added-Variable Plots -- Correlations and Added-Variable Plots -- 2.4.3 Influence Diagnostics -- 2.4.4 Cross-Sectional Correlation -- Testing for Nonzero Cross-Sectional Correlation -- Calibration of Cross-Sectional Correlation Test Statistics -- 2.4.5 Heteroscedasticity -- 2.5 Model Extensions -- 2.5.1 Serial Correlation -- Timing of Observations.

Temporal Covariance Matrix -- 2.5.2 Subject-Specific Slopes -- Sampling and Model Assumptions -- Least-Squares Estimators -- 2.5.3 Robust Estimation of Standard Errors -- Further Reading -- Appendix 2A Least-Squares Estimation -- 2A.1 Basic Fixed-Effects Model: Ordinary Least-Squares Estimation -- 2A.2 Fixed-Effects Models: Generalized Least-Squares Estimation -- 2A.3 Diagnostic Statistics -- Observation-Level Diagnostic Statistic -- Subject-Level Diagnostic Statistic -- 2A.4 Cross-Sectional Correlation: Shortcut Calculations -- Exercises and Extensions -- Section 2.1 -- Section 2.3 -- Section 2.4 -- Section 2.5 -- Empirical Exercises -- 3 Models with Random Effects -- 3.1 Error-Components/Random-Intercepts Model -- Sampling and Inference -- Basic Model and Assumptions -- Traditional ANOVA Setup -- Sampling and Model Assumptions -- Structural Models -- Inference -- Time-Constant Variables -- Degrees of Freedom -- Generalized Least-Squares Estimation -- Feasible GLS Estimator -- Pooling Test -- 3.2 Example: Income Tax Payments -- 3.3 Mixed-Effects Models -- 3.3.1 Linear Mixed-Effects Model -- Special Cases -- Repeated Measures Design -- Random-Coefficients Model -- Variations of the Random-Coefficients Model -- Group Effects -- Time-Constant Variables -- 3.3.2 Mixed Linear Models -- 3.4 Inference for Regression Coefficients -- GLS Estimation -- Matrix Inversion Formula -- Maximum Likelihood Estimation -- Robust Estimation of Standard Errors -- Testing Hypotheses -- 3.5 Variance Components Estimation -- 3.5.1 Maximum Likelihood Estimation -- Iterative Estimation -- 3.5.2 Restricted Maximum Likelihood -- Starting Values -- 3.5.3 MIVQUEs -- Further Reading -- Appendix 3A REML Calculations -- 3A.1 Independence of Residuals and Least-Squares Estimators -- 3A.2 Restricted Likelihoods -- 3A.3 Likelihood Ratio Tests and REML.

Special Case: Testing the Importance of a Subset of Regression Coefficients -- Exercises and Extensions -- Section 3.1 -- Section 3.3 -- Section 3.4 -- Section 3.5 -- Empirical Exercises -- 4 Prediction and Bayesian Inference -- 4.1 Estimators versus Predictors -- 4.2 Predictions for One-Way ANOVA Models -- Shrinkage Estimator -- Best Predictors -- Types of Predictors -- 4.3 Best Linear Unbiased Predictors -- BLUPs as Predictors -- 4.4 Mixed-Model Predictors -- 4.4.1 Linear Mixed-Effects Model -- 4.4.2 Linear Combinations of Global Parameters and Subject-Specific Effects -- 4.4.3 BLUP Residuals -- 4.4.4 Predicting Future Observations -- 4.5 Example: Forecasting Wisconsin Lottery Sales -- 4.5.1 Sources and Characteristics of Data -- 4.5.2 In-Sample Model Specification -- 4.5.3 Out-of-Sample Model Specification -- 4.5.4 Forecasts -- 4.6 Bayesian Inference -- 4.7 Credibility Theory -- 4.7.1 Credibility Theory Models -- 4.7.2 Credibility Rate-Making -- Further Reading -- Appendix 4A Linear Unbiased Prediction -- 4A.1 Minimum Mean-Square Predictor -- 4A.2 Best Linear Unbiased Predictor -- 4A.3 BLUP Variance -- Exercises and Extensions -- Section 4.2 -- Section 4.4 -- Empirical Exercises -- 5 Multilevel Models -- 5.1 Cross-Sectional Multilevel Models -- 5.1.1 Two-Level Models -- Centering of Variables -- Extended Two-Level Models -- Motivation for Multilevel Models -- 5.1.2 Multiple-Level Models -- 5.1.3 Multilevel Modeling in Other Fields -- 5.2 Longitudinal Multilevel Models -- 5.2.1 Two-Level Models -- Growth-Curve Models -- 5.2.2 Multiple-Level Models -- 5.3 Prediction -- Two-Level Models -- Multiple-Level Models -- 5.4 Testing Variance Components -- Further Reading -- Appendix 5A High-Order Multilevel Models -- Exercises and Extensions -- Section 5.3 -- Section 5.4 -- Empirical Exercise -- Appendix 5A -- 6 Stochastic Regressors.

6.1 Stochastic Regressors in Nonlongitudinal Settings -- 6.1.1 Endogenous Stochastic Regressors -- 6.1.2 Weak and Strong Exogeneity -- 6.1.3 Causal Effects -- 6.1.4 Instrumental Variable Estimation -- 6.2 Stochastic Regressors in Longitudinal Settings -- 6.2.1 Longitudinal Data Models without Heterogeneity Terms -- 6.2.2 Longitudinal Data Models with Heterogeneity Terms and Strictly Exogenous Regressors -- Fixed-Effects Estimation -- 6.3 Longitudinal Data Models with Heterogeneity Terms and Sequentially Exogenous Regressors -- Lagged-Dependent Variable Model -- Estimation Difficulties -- 6.4 Multivariate Responses -- 6.4.1 Multivariate Regression -- 6.4.2 Seemingly Unrelated Regressions -- 6.4.3 Simultaneous-Equations Models -- 6.4.4 Systems of Equations with Error Components -- Seemingly Unrelated Regression Models with Error Components -- Simultaneous-Equations Models with Error Components -- 6.5 Simultaneous-Equations Models with Latent Variables -- 6.5.1 Cross-Sectional Models -- Mean Parameters -- Covariance Parameters -- Identification Issues -- Special Cases -- Path Diagrams -- Estimation Techniques -- 6.5.2 Longitudinal Data Applications -- Growth-Curve Models -- Further Reading -- Appendix 6A Linear Projections -- 7 Modeling Issues -- 7.1 Heterogeneity -- Two Approaches to Modeling Heterogeneity -- Practical Identification of Heterogeneity May Be Difficult -- Theoretical Identification with Heterogeneity May Be Impossible -- Estimation of Regression Coefficients without Complete Identification Is Possible -- 7.2 Comparing Fixed- and Random-Effects Estimators -- 7.2.1 A Special Case -- Omitted Variables: Model of Correlated Effects -- 7.2.2 General Case -- Correlated-Effects Model -- 7.3 Omitted Variables -- 7.3.1 Models of Omitted Variables -- 7.3.2 Augmented Regression Estimation -- 7.4 Sampling, Selectivity Bias, and Attrition.

7.4.1 Incomplete and Rotating Panels -- 7.4.2 Unplanned Nonresponse -- Missing-Data Models -- 7.4.3 Nonignorable Missing Data -- Heckman Two-Stage Procedure -- Hausman and Wise Procedure -- EM Algorithm -- Exercises and Extensions -- Section 7.2 -- 8 Dynamic Models -- 8.1 Introduction -- 8.2 Serial Correlation Models -- 8.2.1 Covariance Structures -- 8.2.2 Nonstationary Structures -- 8.2.3 Continuous-Time Correlation Models -- 8.3 Cross-Sectional Correlations and Time-Series Cross-Section Models -- 8.4 Time-Varying Coefficients -- 8.4.1 The Model -- 8.4.2 Estimation -- 8.4.3 Forecasting -- 8.5 Kalman Filter Approach -- 8.5.1 Transition Equations -- 8.5.2 Observation Set -- 8.5.3 Measurement Equations -- 8.5.4 Initial Conditions -- 8.5.5 The Kalman Filter Algorithm -- Likelihood Equations -- 8.6 Example: Capital Asset Pricing Model -- Appendix 8A Inference for the Time-Varying Coefficient Model -- 8A.1 The Model -- 8A.2 Estimation -- Likelihood Equations -- 8A.3 Prediction -- Forecasting -- 9 Binary Dependent Variables -- 9.1 Homogeneous Models -- Linear Probability Models -- 9.1.1 Logistic and Probit Regression Models -- Using Nonlinear Functions of Explanatory Variables -- Threshold Interpretation -- Random-Utility Interpretation -- Logistic Regression -- Odds Ratio Interpretation -- Logistic Regression Parameter Interpretation -- 9.1.2 Inference for Logistic and Probit Regression Models Parameter Estimation -- 9.1.3 Example: Income Tax Payments and Tax Preparers -- 9.2 Random-Effects Models -- Random-Effects Likelihood -- Multilevel Model Extensions -- 9.3 Fixed-Effects Models -- Maximum Likelihood Estimation -- Conditional Maximum Likelihood Estimation -- Conditional Likelihood Estimation -- 9.4 Marginal Models and GEE -- GEE Estimators for the Random-Effects Binary Dependent-Variable Model -- GEE Estimation Procedure -- Further Reading.

Appendix 9A Likelihood Calculations.