Missing Data, Issue 136 (Google eBook)Using numerous examples and practical tips, this book offers a nontechnical explanation of the standard methods for missing data (such as listwise or casewise deletion) as well as two newer (and, better) methods, maximum likelihood and multiple imputation. Anyone who has relied on adhoc methods that are statistically inefficient or biased will find this book a welcome and accessible solution to their problems with handling missing data.

What people are saying  Write a review
We haven't found any reviews in the usual places.
Contents
Introduction  1 
Assumptions  3 
Missing at Random  4 
Ignorable  5 
Listwise Deletion  6 
Pairwise Deletion  8 
Dummy Variable Adjustment  9 
Imputation  11 
Sequential Versus Parallel Chains of Data Augmentation  37 
Using the Normal Model for Nonnormal or Categorical Data  38 
Exploratory Analysis  41 
Multiple Imputation Complications  50 
Compatibility of the Imputation Model and the Analysis Model  52 
Role of the Dependent Variable in Imputation  53 
Using Additional Variables in the Imputation Process  54 
Other Parametric Approaches to Multiple Imputation  55 
Summary  12 
Review of Maximum Likelihood  13 
ML With Missing Data  14 
Contingency Table Data  15 
Linear Models With Normally Distributed Data  18 
The EM Algorithm  19 
EM Example  21 
Direct ML  23 
Direct ML Example  25 
Conclusion  26 
Multiple Imputation Basics  27 
Single Random Imputation  28 
Multiple Random Imputation  29 
Allowing for Random Variation in the Parameter Estimates  30 
Multiple Imputation Under the Multivariate Normal Model  32 
Data Augmentation for the Multivariate Normal Model  34 
Convergence in Data Augmentation  36 
Nonparametric and Partially Parametric Methods  57 
Sequential Generalized Regression Models  64 
Linear Hypothesis Tests and Likelihood Ratio Tests  65 
MI Example 2  68 
MI for Longitudinal and Other Clustered Data  73 
MI Example 3  74 
Nonignorable Missing Data  77 
Two Classes of Models  78 
Heckmans Model for Sample Selection Bias  79 
ML Estimation With PatternMixture Models  82 
Multiple Imputation With PatternMixture Models  83 
Summary and Conclusion  84 
Notes  87 
References  89 
About the Author  93 
Common terms and phrases
algorithm applied approach biased calculated chisquare completed data sets completely at random compute contingency table convergence correlation covariance matrix data are missing degrees of freedom dependent variable Direct ML dummy variable dummy variable adjustment EM algorithm equation esti example five data sets formula imputation methods imputation process imputed data imputed values independent likelihood function likelihood ratio linear models linear regression listwise deletion loglinear models logistic regression maximum likelihood missing at random missing completely missing data mechanism missing data patterns missing information missing values missingness ML estimates multiple imputation multivariate normal model nonignorable missing data normal distribution observed pairwise deletion parameter estimates patternmixture models Poisson regression posterior distribution predicted values predictors probability of missing produce random draws Random Imputation randomly regression coefficients regression model residual Rubin sample Schafer sequential SPANKING standard error estimates standard errors starting values Suppose tion variables with missing variance