This book provides a modern approach to a number of statistical model concepts that are linear in the unknown parameters and include a random error term. Aresult of the author's 30 years of practical and teaching experience, this Second Edition follows an introduction-theorem-proof-examples format that allowsfor easier comprehension of how to use the methods and also how to recognize the associated assumptions and limits. In addition to discussions on the methods of study of random vectors, multiple regression techniques, simultaneous confidence intervals, and analysis of frequency data, new topics such as mixed models, nonparametric regression, Bayesian models, multiple comparison procedures, estimation of quantiles for regression models, and curve fitting of modelshave been added in an effort to thoroughly update and modernize the book. This new edition emphasizes the geometry of vector spaces (the coordinate free approach) because the author has found that the intuition it provides is vital to the understanding of the theory. Historical notes can be found throughout the book, and it also includes examples of applications using S-Plus, R, and SASpackages.
- ISBN: 978-0-470-23146-3
- Editorial: John Wiley & Sons
- Encuadernacion: Rústica
- Páginas: 512
- Fecha Publicación: 24/07/2009
- Nº Volúmenes: 1
- Idioma: Inglés