Semiparametric Regression for the Social Sciences

摘要: List of Tables. Figures. Preface. 1 Introduction: Global versus Local Statistics. 1.1 The Consequences Ignoring Nonlinearity. 1.2 Power Transformations. 1.3 Nonparametric and Semiparametric Techniques. 1.4 Outline the Text. 2 Smoothing Regression. 2.1 Simple Smoothing. 2.1.1 Averaging. 2.1.2 Kernel 2.2 Polynomial 2.3 Modeling Choices. 2.3.1 Span. 2.3.2 Degree Weight Function. 2.3.3 A Note on Interpretation. 2.4 Statistical Inference for 2.5 Multiple 2.6 Conclusion. 2.7 Exercises. 3 Splines. 3.1 Regression 3.1.1 Basis Functions. 3.2 Other Spline Models Bases. 3.2.1 Quadratic Cubic 3.2.2 Natural 3.2.3 B-splines. 3.2.4 Knot Placement Numbers. 3.2.5 Comparing Models. 3.3 Splines Overfitting. 3.3.1 3.3.2 as Mixed 3.3.3 Final Notes 3.3.4 Thin Plate 3.4 3.5 Comparisons Conclusions. 3.6 4 Automated 4.1 Span by Cross-Validation. 4.2 4.2.1 Estimating Through Likelihood. 4.2.2 4.3 in Practice. 4.4 Caveats. 4.5 5 Additive 5.1 5.2 5.3 Estimation. 5.3.1 Backfitting. 5.4 Inference. 5.5 Examples. 5.5.1 Congressional Elections. 5.5.2 Feminist Attitudes. 5.6 Discussion. 5.7 6 Generalized 6.1 Linear 6.2 Estimation GAMS. 6.3 6.4 6.4.1 Logistic Regression: Liberal Peace. 6.4.2 Ordered Logit: Domestic Violence. 6.4.3 Count Models: Supreme Court Overrides. 6.4.4 Survival Race Riots. 6.5 6.6 7 Extensions Model. 7.1 7.2 Bayesian 7.3 Propensity Score Matching. 7.4 8 Bootstrapping. 8.1 Classical 8.2 Bootstrapping - An Overview. 8.2.1 8.2.2 Example: Mean. 8.2.3 8.2.4 Presidential 8.3 8.3.1 Fits. 8.3.2 Nonlinearity Tests. 8.4 8.5 9 Epilogue. Appendix: Software. Bibliography. Author Index. Subject