Statistical Methods for Handling Incomplete Data
By Jae Kwang Kim, Jun Shao
To Be Published July 17th 2013 by Chapman and Hall/CRC – 215 pages
Description
With advances in computing power, there have been substantial developments in computational methods for handling missing data. This text presents an introduction to the theory, applications, and computational aspects of missing data analysis. It covers the three main methodological approaches: likelihood-based, nonparametric, and quasi-randomization. The text includes many real examples and integrates computer code where appropriate. It also provides exercises at the end of each chapter. A solutions manual is available for qualifying instructors.
Contents
Introduction
Likelihood-Based Approach
Introduction
Observed Likelihood
Mean Score Approach
Fisher Information in the Observed Likelihood
Computation for MLE
Introduction
Factoring Likelihood Approach
EM Algorithm
Monte Carlo Approach
Introduction
Monte Carlo EM Algorithm
Data Augmentation
Imputation
Introduction
Basic Theory for Imputation
Multiple Imputation
Fractional Imputation
Pseudo ML Approach
Introduction
Theory for Pseudo ML Estimation
Nonparametric Approach
Introduction
Kernel Method
Empirical Likelihood Method
Nonparametric Imputation
Quasi-Randomization Approach
Introduction
Propensity Score Method
Doubly Robust Method
Nonignorable Missing Data
Longitudinal Missing Data
Other Topics
Name: Statistical Methods for Handling Incomplete Data (Hardback) – Chapman and Hall/CRC
Description: By Jae Kwang Kim, Jun Shao. With advances in computing power, there have been substantial developments in computational methods for handling missing data. This text presents an introduction to the theory, applications, and computational aspects of missing data analysis. It covers...
Categories: Statistical Theory & Methods, Statistics & Computing, Statistics for the Biological Sciences
