A novel approach to linear model analysis
Methods and Applications of Linear Models provides a clear and concise summary of the concepts and methodologies of linear models and illustrates the analysis with numerous exercises and real-world examples. Special features include:
- Data sets available on an ftp site
- Graphical illustrations of many of the analyses
- A data-based approach to development and analysis
- Graphical and numerical diagnostic methods in regression
- Use of the cell means model for the analysis of variance
- The introduction of the AVE method for variance component estimation
- A general approach to the analysis of unbalanced mixed models
This novel approach to linear model analysis offers a unified treatment of linear regression and the analysis of variance. The focus is on the appropriate interpretation of results. Carefully chosen examples illustrate the analyses and some of the common sources of confusion in the application of the methods. The treatment of mixed models includes material that has not previously appeared in the literature.
For upper-level undergraduate and graduate students of regression and the analysis of variance, this volume provides simple explanations of the basic methodologies. It is also a valuable professional reference for applied statisticians and researchers. Ronald R. Hocking is Professor Emeritus in the Department of Statistics at Texas A&M University. He received his PhD in mathematics and statistics and is a Fellow of the American Statistical Association.
Regression and analysis of variance represent 90 of all applied statistical analysis. This book is unique in that it represents a unified treatment of these two areas. This view is carried out through chapters discussing linear models, distribution of linear and quadratic forms, estimation and inference for simple linear models, single predictor regression models, multiple predictor regression models as well as factorial models. A disk of data sets will be included in the book.
