Foundations of Computational Statistics (Photocopy Edition)

Author: James E. Gentle
Publisher:
Publication Date: 2006-01-17
Features: The method of intensive computation has been widely used in statistical inference and exploratory data analysis. Computational statistics methods include resampling, classification, and multiple transformations of datasets, which may involve the use of artificially generated random data. The application of these methods requires advanced techniques in numerical analysis. Therefore, computational statistics and statistical computing methods are closely related. This book elaborates on various methods of computational statistics and some applications of the intensive computation method in areas such as density estimation, structural confirmation of data, and model building. Although this book does not specifically discuss statistical computing methods, it comprehensively covers numerical techniques in data transformation, function approximation, and data optimization under the meaning of statistical methods. The book provides exercises, some of which include solutions. While assuming readers are familiar with probability theory and statistics, it also reviews the basic methods of statistical inference, making the book largely self-contained. It can be used as a textbook or supplementary text for modern statistics courses for senior undergraduate or graduate students, or as a reference book for statisticians using intensive computation methods. The author is a professor in the Department of Computational Statistics at George Mason University, a member of the American Statistical Association (ASA) and the International Statistical Society, and holds multiple national offices in ASA. He is an associate editor of ASA journals and serves as an editor for other journals in statistics and computing. He is also the author of "Random Number Generation, Monte Carlo Methods, and Numerical Linear Algebra in Statistics."