Regression with R
Preface
The book is undergoing conversion from an older version and format. It is currently incomplete and subject to change without notice.
The state of individual chapters is marked at the top of the chapter.
This book was written as the textbook material for a graduate statistics course in regression analysis. The prerequisites include univariate linear regression, \(t\)-tests and constructions of standard confidence intervals, and knowledge of standard distributions such as the normal, the binomial and the Poisson distribution. The reader will also benefit from introductory statistics courses covering likelihood methods, one- and two-sided analysis of variance, and aspects of asymptotic theory. In addition, a solid knowledge of linear algebra is assumed.
The exposition is mathematical, but the emphasis is on data modeling as well as theory. Mathematics is used to make model descriptions and assumptions precise, and to give precise results relevant for the practical analysis of data and the computations required for carrying out data analysis.
The book attempts to be complete and thorough on the topics covered, yet to be practical and relevant for the applied statistician. The means for achieving the latter is by larger case studies using R. The R code included is complete and covers most aspects of the data analysis from reading data into R, cleaning and plotting data to data analysis, data modeling and model diagnostics.
The book source is available from the RwR GitHub repository and it is accompanied by the R package RwR containing datasets used in the book.
The book is undergoing conversion from an older version and format. It is currently incomplete and subject to change without notice.
The state of individual chapters is marked at the top of the chapter.