The provided text is the preface and introduction of a book titled "Analysis of Incomplete Multivariate Data" by Joseph L. Schafer. The book aims to provide practical tools for handling missing data in multivariate datasets, focusing on datasets where missing values may occur on any or all variables. The authors emphasize the importance of using principled methods that account for missing data and the uncertainty they introduce, rather than ad hoc methods like case deletion or imputation. The book covers a range of topics, including: 1. **Introduction**: Discusses the purpose, background, and why simulation-based analysis is preferred over ad hoc methods. 2. **Background**: Explains the EM algorithm and Markov chain Monte Carlo (MCMC) techniques, which are fundamental to the book's approach. 3. **Practical Inference**: Describes how to conduct inference by simulation, including parameter simulation and multiple imputation. 4. **Algorithms for Specific Models**: Details algorithms for specific multivariate models, such as the multivariate normal distribution, cross-classified categorical data, and mixed datasets. 5. **Software and Computational Details**: Provides information on implementing the algorithms in the statistical language S and encourages readers to adapt them to other environments. The book is intended for applied statisticians, graduate students, and researchers who need practical tools for handling missing data. It emphasizes computational efficiency and practicality, making it suitable for a wide range of applications. The authors also provide references and bibliographic notes for further reading on the topics covered.The provided text is the preface and introduction of a book titled "Analysis of Incomplete Multivariate Data" by Joseph L. Schafer. The book aims to provide practical tools for handling missing data in multivariate datasets, focusing on datasets where missing values may occur on any or all variables. The authors emphasize the importance of using principled methods that account for missing data and the uncertainty they introduce, rather than ad hoc methods like case deletion or imputation. The book covers a range of topics, including: 1. **Introduction**: Discusses the purpose, background, and why simulation-based analysis is preferred over ad hoc methods. 2. **Background**: Explains the EM algorithm and Markov chain Monte Carlo (MCMC) techniques, which are fundamental to the book's approach. 3. **Practical Inference**: Describes how to conduct inference by simulation, including parameter simulation and multiple imputation. 4. **Algorithms for Specific Models**: Details algorithms for specific multivariate models, such as the multivariate normal distribution, cross-classified categorical data, and mixed datasets. 5. **Software and Computational Details**: Provides information on implementing the algorithms in the statistical language S and encourages readers to adapt them to other environments. The book is intended for applied statisticians, graduate students, and researchers who need practical tools for handling missing data. It emphasizes computational efficiency and practicality, making it suitable for a wide range of applications. The authors also provide references and bibliographic notes for further reading on the topics covered.