| |
| |
Preface | |
| |
| |
| |
Overview and Basic Approaches | |
| |
| |
| |
Introduction | |
| |
| |
| |
The Problem of Missing Data | |
| |
| |
| |
Missing-Data Patterns | |
| |
| |
| |
Mechanisms That Lead to Missing Data | |
| |
| |
| |
A Taxonomy of Missing-Data Methods | |
| |
| |
| |
Missing Data in Experiments | |
| |
| |
| |
Introduction | |
| |
| |
| |
The Exact Least Squares Solution with Complete Data | |
| |
| |
| |
The Correct Least Squares Analysis with Missing Data | |
| |
| |
| |
Filling in Least Squares Estimates | |
| |
| |
| |
Bartlett's ANCOVA Method | |
| |
| |
| |
Least Squares Estimates of Missing Values by ANCOVA Using Only Complete-Data Methods | |
| |
| |
| |
Correct Least Squares Estimates of Standard Errors and One Degree of Freedom Sums of Squares | |
| |
| |
| |
Correct Least Squares Sums of Squares with More Than One Degree of Freedom | |
| |
| |
| |
Complete-Case and Available-Case Analysis, Including Weighting Methods | |
| |
| |
| |
Introduction | |
| |
| |
| |
Complete-Case Analysis | |
| |
| |
| |
Weighted Complete-Case Analysis | |
| |
| |
| |
Available-Case Analysis | |
| |
| |
| |
Single Imputation Methods | |
| |
| |
| |
Introduction | |
| |
| |
| |
Imputing Means from a Predictive Distribution | |
| |
| |
| |
Imputing Draws from a Predictive Distribution | |
| |
| |
| |
Conclusions | |
| |
| |
| |
Estimation of Imputation Uncertainty | |
| |
| |
| |
Introduction | |
| |
| |
| |
Imputation Methods that Provide Valid Standard Errors from a Single Filled-in Data Set | |
| |
| |
| |
Standard Errors for Imputed Data by Resampling | |
| |
| |
| |
Introduction to Multiple Imputation | |
| |
| |
| |
Comparison of Resampling Methods and Multiple Imputation | |
| |
| |
| |
Likelihood-Based Approaches to the Analysis of Missing Data | |
| |
| |
| |
Theory of Inference Based on the Likelihood Function | |
| |
| |
| |
Review of Likelihood-Based Estimation for Complete Data | |
| |
| |
| |
Likelihood-Based Inference with Incomplete Data | |
| |
| |
| |
A Generally Flawed Alternative to Maximum Likelihood: Maximizing Over the Parameters and the Missing Data | |
| |
| |
| |
Likelihood Theory for Coarsened Data | |
| |
| |
| |
Factored Likelihood Methods, Ignoring the Missing-Data Mechanism | |
| |
| |
| |
Introduction | |
| |
| |
| |
Bivariate Normal Data with One Variable Subject to Nonresponse: ML Estimation | |
| |
| |
| |
Bivariate Normal Monotone Data: Small-Sample Inference | |
| |
| |
| |
Monotone Data With More Than Two Variables | |
| |
| |
| |
Factorizations for Special Nonmonotone Patterns | |
| |
| |
| |
Maximum Likelihood for General Patterns of Missing Data: Introduction and Theory with Ignorable Nonresponse | |
| |
| |
| |
Alternative Computational Strategies | |
| |
| |
| |
Introduction to the EM Algorithm | |
| |
| |
| |
The E and M Steps of EM | |
| |
| |
| |
Theory of the EM Algorithm | |
| |
| |
| |
Extensions of EM | |
| |
| |
| |
Hybrid Maximization Methods | |
| |
| |
| |
Large-Sample Inference Based on Maximum Likelihood Estimates | |
| |
| |
| |
Standard Errors Based on the Information Matrix | |
| |
| |
| |
Standard Errors via Methods that do not Require Computing and Inverting an Estimate of the Observed Information Matrix | |
| |
| |
| |
Bayes and Multiple Imputation | |
| |
| |
| |
Bayesian Iterative Simulation Methods | |
| |
| |
| |
Multiple Imputation | |
| |
| |
| |
Likelihood-Based Approaches to the Analysis of Incomplete Data: Some Examples | |
| |
| |
| |
Multivariate Normal Examples, Ignoring the Missing-Data Mechanism | |
| |
| |
| |
Introduction | |
| |
| |
| |
Inference for a Mean Vector and Covariance Matrix with Missing Data Under Normality | |
| |
| |
| |
Estimation with a Restricted Covariance Matrix | |
| |
| |
| |
Multiple Linear Regression | |
| |
| |
| |
A General Repeated-Measures Model with Missing Data | |
| |
| |
| |
Time Series Models | |
| |
| |
| |
Robust Estimation | |
| |
| |
| |
Introduction | |
| |
| |
| |
Robust Estimation for a Univariate Sample | |
| |
| |
| |
Robust Estimation of the Mean and Covariance Matrix | |
| |
| |
| |
Further Extensions of the t Model | |
| |
| |
| |
Models for Partially Classified Contingency Tables, Ignoring the Missing-Data Mechanism | |
| |
| |
| |
Introduction | |
| |
| |
| |
Factored Likelihoods for Monotone Multinomial Data | |
| |
| |
| |
ML and Bayes Estimation for Multinomial Samples with General Patterns of Missing Data | |
| |
| |
| |
Loglinear Models for Partially Classified Contingency Tables | |
| |
| |
| |
Mixed Normal and Non-normal Data with Missing Values, Ignoring the Missing-Data Mechanism | |
| |
| |
| |
Introduction | |
| |
| |
| |
The General Location Model | |
| |
| |
| |
The General Location Model with Parameter Constraints | |
| |
| |
| |
Regression Problems Involving Mixtures of Continuous and Categorical Variables | |
| |
| |
| |
Further Extensions of the General Location Model | |
| |
| |
| |
Nonignorable Missing-Data Models | |
| |
| |
| |
Introduction | |
| |
| |
| |
Likelihood Theory for Nonignorable Models | |
| |
| |
| |
Models with Known Nonignorable Missing-Data Mechanisms: Grouped and Rounded Data | |
| |
| |
| |
Normal Selection Models | |
| |
| |
| |
Normal Pattern-Mixture Models | |
| |
| |
| |
Nonignorable Models for Normal Repeated-Measures Data | |
| |
| |
| |
Nonignorable Models for Categorical Data | |
| |
| |
References | |
| |
| |
Author Index | |
| |
| |
Subject Index | |