[R] Stat textbook recommendations?

[Peter Dalgaard]

(resending to include r-help)

Monte Milanuk wrote:
> Hello,
> 
> I'm looking for a textbook that can explain some of the math behind  
> the intro-to-intermediate stuff like ANOVA, multiple regression, non- 
> parametric tests, etc.
> 
> A little background:  I took an intro stats course last year and  
> would like to further my education.  Being as that was the highest  
> (and only) stats class the local community college offers, it looks  
> like I'm on my own from here.  I've been working through some of the  
> online 'stats with R' tutorials as well as Dalgaard's ISWR.  Where  
> I'm running into problems is the transition from Bluman's 'A Brief  
> Introduction to Elementary Statistics' (covers up through paired t- 
> tests, chi-squared/goodness-of-fit, simple linear regression &  
> correlation, and just barely mentions ANOVA) with a TI-83+, to even  
> books like ISWR... when they start getting into the things like one  
> and two-way ANOVA, multiple regression, model selection, survival,  
> etc. I start feeling like I have one hand tied behind my back - I  
> just don't have enough theoretical exposure to really understand what  
> techniques I would use when, relative to my own projects outside the  
> book.
> 
> Several of the 'intro to stats using R' books and pdf tutorials  
> mention that they are not really meant as a standalone statistics  
> text book, but in addition to a traditional stats textbook (Verzani  
> mentions Kitchen's book specifically).  So I guess what I'm looking  
> for is any other recommendations on intro or intermediate textbooks  
> that deal primarily with the math/theory behind the processes.  If  
> they were oriented towards R that's be great, but otherwise I guess  
> I'd be most interested in something relatively platform-agnostic -  
> I've seen some books that were slanted heavily towards a particular  
> software package (Minitab) that I cannot afford or justify for  
> personal use.

Re. ISwR, you might want to take notice that it was originally written
for a course that used Altman's "Practical Statistics for Medical
Research". It is, however, a bit wordy for some and glosses rather too
quickly over the math.

Another popular item for ambitious beginners is Kirkwood and Sterne:
Essential Medical Statistics. Their notation is a bit maddening (for
teachers anyway) but they do cover a lot of ground without digging too
deeply into the math.

If you want more math, beware that what is good, strongly depends on
your prerequisites. Linear model theory, e.g., gets much easier with
matrix calculus and nearly trivial if you know about abstract linear
algebra and projections in N dimensional vector spaces. For relatively
basic levels, look at booke that are popular for first courses in
Engineering: Devore, Johnson+Miller+Freund, and probably more.

-- 
    O__  ---- Peter Dalgaard             Øster Farimagsgade 5, Entr.B
   c/ /'_ --- Dept. of Biostatistics     PO Box 2099, 1014 Cph. K
  (*) \(*) -- University of Copenhagen   Denmark      Ph:  (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)              FAX: (+45) 35327907