Statistical Methods for Handling Incomplete Data

Due to recent theoretical findings and advances in statistical computing, there has been a rapid development of techniques and applications in the area of missing data analysis. Statistical Methods for Handling Incomplete Data covers the most up-to-date statistical theories and computational methods for analyzing incomplete data.Suitable for graduate students and researchers in statistics, the book presents thorough treatments of: Statistical theories of likelihood-based inference with missing dataComputational techniques and theories on imputationMethods involving propensity score weighting, nonignorable missing data, longitudinal missing data, survey sampling, and statistical matchingAssuming prior experience with statistical theory and linear models, the text uses the frequentist framework with less emphasis on Bayesian methods and nonparametric methods. It includes many examples to help readers understand the methodologies. Some of the research ideas introduced can be developed further for specific applications.Table of ContentIntroduction. Likelihood-Based Approach. Computation. Imputation. Propensity Scoring Approach. Nonignorable Missing Data. Longitudinal and Clustered Data. Application to Survey Sampling. Statistical Matching. Bibliography. Index.Review Quote"… this book nicely blends the theoretical material and its application through examples, and will be of interest to students and researchers as a textbook or a reference book. Extensive coverage of recent advances in handling missing data provides resources and guidelines for researchers and practitioners in implementing the methods in new settings. … I plan to use this as a textbook for my teaching and highly recommend it."—Biometrics, September 2014