| Preface |
| Casual inference and observational studies / I： |
| An overview of methods for causal inference from observational studies / Sander Greenland1： |
| Introduction / 1.1： |
| Approaches based on causal models / 1.2： |
| Canonical inference / 1.3： |
| Methodologic modeling / 1.4： |
| Conclusion / 1.5： |
| Matching in observational studies / Paul R. Rosenbaum2： |
| The role of matching in observational studies / 2.1： |
| Why match? / 2.2： |
| Two key issues: balance and structure / 2.3： |
| Additional issues / 2.4： |
| Estimating causal effects in nonexperimental studies / Rajeev Dehejia3： |
| Identifying and estimating the average treatment effect / 3.1： |
| The NSWdata / 3.3： |
| Propensity score estimates / 3.4： |
| Conclusions / 3.5： |
| Medication cost sharing and drug spending in Medicare / Alyce S. Adams4： |
| Methods / 4.1： |
| Results / 4.2： |
| Study limitations / 4.3： |
| Conclusions and policy implications / 4.4： |
| A comparison of experimental and observational data analyses / Jennifer L. Hill ; Jerome P. Reiter ; Elaine L. Zanutto5： |
| Experimental sample / 5.1： |
| Constructed observational study / 5.2： |
| Concluding remarks / 5.3： |
| Fixing broken experiments using the propensity score / Bruce Sacerdote6： |
| The lottery data / 6.1： |
| Estimating the propensity scores / 6.3： |
| The propensity score with continuous treatments / Keisuke Hirano ; Guido W. Imbens6.4： |
| The basic framework / 7.1： |
| Bias removal using the GPS / 7.3： |
| Estimation and inference / 7.4： |
| Application: the Imbens-Rubin-Sacerdote lottery sample / 7.5： |
| Causal inference with instrumental variables / Junni L. Zhang7.6： |
| Key assumptions for the LATE interpretation of the IV estimand / 8.1： |
| Estimating causal effects with IV / 8.3： |
| Some recent applications / 8.4： |
| Discussion / 8.5： |
| Principal stratification / Constantine E. Frangakis9： |
| Introduction: partially controlled studies / 9.1： |
| Examples of partially controlled studies / 9.2： |
| Estimands / 9.3： |
| Assumptions / 9.5： |
| Designs and polydesigns / 9.6： |
| Missing data modeling / II： |
| Nonresponse adjustment in government statistical agencies: constraints, inferential goals, and robustness issues / John L. Eltinge10： |
| Introduction: a wide spectrum of nonresponse adjustment efforts in government statistical agencies / 10.1： |
| Constraints / 10.2： |
| Complex estimand structures, inferential goals, and utility functions / 10.3： |
| Robustness / 10.4： |
| Closing remarks / 10.5： |
| Bridging across changes in classification systems / Nathaniel Schenker11： |
| Multiple imputation to achieve comparability of industry and occupation codes / 11.1： |
| Bridging the transition from single-race reporting to multiple-race reporting / 11.3： |
| Representing the Census undercount by multiple imputation of households / Alan M. Zaslavsky11.4： |
| Models / 12.1： |
| Inference / 12.3： |
| Simulation evaluations / 12.4： |
| Statistical disclosure techniques based on multiple imputation / Roderick J. A. Little ; Fang Liu ; Trivellore E. Raghunathan12.5： |
| Full synthesis / 13.1： |
| SMIKe and MIKe / 13.3： |
| Analysis of synthetic samples / 13.4： |
| An application / 13.5： |
| Designs producing balanced missing data: examples from the National Assessment of Educational Progress / Neal Thomas13.6： |
| Statistical methods in NAEP / 14.1： |
| Split and balanced designs for estimating population parameters / 14.3： |
| Maximum likelihood estimation / 14.4： |
| The role of secondary covariates / 14.5： |
| Propensity score estimation with missing data / Ralph B. D Agostino Jr14.6： |
| Notation / 15.1： |
| Applied example:March of Dimes data / 15.3： |
| Conclusion and future directions / 15.4： |
| Sensitivity to nonignorability in frequentist inference / Guoguang Ma ; Daniel F. Heitjan16： |
| Missing data in clinical trials / 16.1： |
| Ignorability and bias / 16.2： |
| A nonignorable selection model / 16.3： |
| Sensitivity of the mean and variance / 16.4： |
| Sensitivity of the power / 16.5： |
| Sensitivity of the coverage probability / 16.6： |
| An example / 16.7： |
| Statistical modeling and computation / 16.8： |
| Statistical modeling / 17： |
| The NSW data |
| Applied example: March of Dimes data / Ralph B. D'Agostino Jr. |
| Regression models / D. Michael Titterington17.1： |
| Latent-variable problems / 17.2： |
| Computation: non-Bayesian / 17.3： |
| Computation: Bayesian / 17.4： |
| Prospects for the future / 17.5： |
| Treatment effects in before-after data / Andrew Gelman18： |
| Default statistical models of treatment effects / 18.1： |
| Before-after correlation is typically larger for controls than for treated units / 18.2： |
| A class of models for varying treatment effects / 18.3： |
| Multimodality in mixture models and factor models / Eric Loken18.4： |
| Multimodality in mixture models / 19.1： |
| Multimodal posterior distributions in continuous latent variable models / 19.2： |
| Summary / 19.3： |
| Modeling the covariance and correlation matrix of repeated measures / W. John Boscardin ; Xiao Zhang20： |
| Modeling the covariance matrix / 20.1： |
| Modeling the correlation matrix / 20.3： |
| Modeling a mixed covariance-correlation matrix / 20.4： |
| Nonzero means and unbalanced data / 20.5： |
| Multivariate probit model / 20.6： |
| Example: covariance modeling / 20.7： |
| Example: mixed data / 20.8： |
| Robit regression: a simple robust alternative to logistic and probit regression / Chuanhai Liu21： |
| The robit model / 21.1： |
| Robustness of likelihood-based inference using logistic, probit, and robit regression models / 21.3： |
| Complete data for simple maximum likelihood estimation / 21.4： |
| Maximum likelihood estimation using EM-type algorithms / 21.5： |
| A numerical example / 21.6： |
| Using EM and data augmentation for the competing risks model / Radu V. Craiu ; Thierry Duchesne21.7： |
| The model / 22.1： |
| EM-based analysis / 22.3： |
| Bayesian analysis / 22.4： |
| Example / 22.5： |
| Discussion and further work / 22.6： |
| Mixed effects models and the EM algorithm / Florin Vaida ; Xiao-Li Meng ; Ronghui Xu23： |
| Binary regression with random effects / 23.1： |
| Proportional hazards mixed-effects models / 23.3： |
| The sampling/importance resampling algorithm / Kim-Hung Li24： |
| SIR algorithm / 24.1： |
| Selection of the pool size / 24.3： |
| Selection criterion of the importance sampling distribution / 24.4： |
| The resampling algorithms / 24.5： |
| Applied Bayesian inference / 24.6： |
| Whither applied Bayesian inference? / Bradley P. Carlin25： |
| Where we've been / 25.1： |
| Where we are / 25.2： |
| Where we're going / 25.3： |
| Efficient EM-type algorithms for fitting spectral lines in high-energy astrophysics / David A. van Dyk ; Taeyoung Park26： |
| Application-specific statistical methods / 26.1： |
| The Chandra X-ray observatory / 26.2： |
| Fitting narrow emission lines / 26.3： |
| Model checking and model selection / 26.4： |
| Improved predictions of lynx trappings using a biological model / Cavan Reilly ; Angelique Zeringue27： |
| The current best model / 27.1： |
| Biological models for predator prey systems / 27.3： |
| Some statistical models based on the Lotka-Volterra system / 27.4： |
| Computational aspects of posterior inference / 27.5： |
| Posterior predictive checks and model expansion / 27.6： |
| Prediction with the posterior mode / 27.7： |
| Record linkage using finite mixture models / Michael D. Larsen27.8： |
| Introduction to record linkage / 28.1： |
| Record linkage / 28.2： |
| Mixture models / 28.3： |
| Application / 28.4： |
| Analysis of linked files / 28.5： |
| Bayesian hierarchical record linkage / 28.6： |
| Identifying likely duplicates by record linkage in a survey of prostitutes / Thomas R. Belin ; Hemant Ishwaran ; Naihua Duan ; Sandra H. Berry ; David E. Kanouse28.7： |
| Concern about duplicates in an anonymous survey / 29.1： |
| General frameworks for record linkage / 29.2： |
| Estimating probabilities of duplication in the Los Angeles Women's Health Risk Study / 29.3： |
| Applying structural equation models with incomplete data / Hal S. Stern ; Yoonsook Jeon29.4： |
| Structural equation models / 30.1： |
| Bayesian inference for structural equation models / 30.2： |
| Iowa Youth and Families Project example / 30.3： |
| Summary and discussion / 30.4： |
| Perceptual scaling / Ying Nian Wu ; Cheng-En Guo ; Song Chun Zhu31： |
| Sparsity and minimax entropy / 31.1： |
| Complexity scaling law / 31.3： |
| Perceptibility scaling law / 31.4： |
| Texture = imperceptible structures / 31.5： |
| Perceptibility and sparsity / 31.6： |
| References |
| Index |
| Preface |
| Casual inference and observational studies / I： |
| An overview of methods for causal inference from observational studies / Sander Greenland1： |
ＥＢ
