-
PDF
- Split View
-
Views
-
Cite
Cite
Fernando Tusell, Linear Models with R, Journal of the Royal Statistical Society Series A: Statistics in Society, Volume 168, Issue 3, July 2005, Page 632, https://doi.org/10.1111/j.1467-985X.2005.00368_5.x
Close - Share Icon Share
This book adds to the growing literature documenting the statistical and graphics system R (see R Development Core Team (2004)), an open source implementation of the language S. It covers linear regression and analysis of variance, with a wealth of examples illustrating estimation, inference, diagnostics, departures from the usual assumptions, variable selection, transformation of variables and the basics of analysis of variance and experimental design. It also includes two short appendix chapters on obtaining and installing R and a quick introduction to the language.
One danger with applied books such as this is that they become recipe lists of the kind ‘press this key to get that result’. This is not so with Faraway's book. Throughout, it gives plenty of insight on what is going on, with comments that even the seasoned practitioner will appreciate. Interspersed with R code and the output that it produces one can find many little gems of what I think is sound statistical advice, well epitomized with the examples chosen.
The book is well produced with few misprints, none of which I found particularly insidious or irritating.
A good book such as this begs the question of who are its potential readers. As is clear from my comments above, I read it with delight and think that the same will be true with anyone who is engaged in the use or teaching of linear models. However, when I tried to read it with the eyes of my students, I found that most of them would find it much too terse. The author acknowledges that the book is not an introduction to R, which is true: a reader with no previous exposure to R might be better served by books such as Dalgaard (2002) or Venables and Ripley (1999) (abundant documentation for R is also available on line). The book also assumes some acquaintance with the theory of linear models. It is quite adequate as a companion book to any of the many standard texts such as Seber (1977), Draper and Smith (1981) or Stapleton (1995), but I think that a student who has no previous training in linear models would find Faraway's discussion of the topic difficult to follow at times. In that particular aspect, Fox (2002) benefits from its tighter integration with a text on regression.
I find this book a valuable buy for anyone who is involved with R and linear models, and it is essential in any university library where those topics are taught. Some more pointers to standard regression texts and an index listing functions by name for easy reference would be on my wish list for future editions.