[R] Fish and 'must read' statistics books. My last word, promise!

Bill.Venables@cmis.csiro.au Bill.Venables at cmis.csiro.au
Sat May 18 08:07:17 CEST 2002

I seem to have unwittingly stirred up a hornets' nest here, so let me have
one more say and leave it.

Firstly, congratulations to Renaud Launcelot, who seems to have a better
feel for wryness and irony in English than some of my native speaker
colleagues.  Of course R A Fisher is a seminal writer, indeed close to the
founder of modern statistics and every aspiring statistician should read
what he had to say _very_ carefully, but don't expect an easy ride.  He had
a deceptively simple style, but very polemical and rather authoritative in a
way that often wrankles with modern readers.  At first reading you may think
you have grasped the meaning but, (if you are anything like me), at the next
reading you will start to realise that the message is deeper, more subtle
and more intriguing than you thought possible.  And the same the next time,
and the next.  Reading Fisher just once is probably not a good idea, in
fact.  0 times is not a disaster, but 4 or 5 times is probably getting close
to maximum utility.  I could hardly imagine a writer with ideas more
inimical to those of Fisher than Jimmie Savage and yet his last
(posthumously published) paper was entitled 'On re-reading R A Fisher'
(note: RE-reading) and it was very complimentary to his old adversary.  Even
for Jimmy the penny dropped, somewhat, in the end.  [OK, there is still a
bit of overstatement and irony in there, but you are on your own from here
on in.]

As far as 'must read' books for Statistics goes for me there are none, but
my all-time favourite statistical author is definitely Sir David Cox.  His
writing is clear and lucid, his ideas are usually immensely powerful,
perceptive and often more general than the context strictly requires.  I'm
not suggesting that everything is perfect or even easy, but even when things
are a bit disorganised and some jagged edges are showing (for example in the
1971 book with Hinkley) you can learn something from the way things appear
to be still developing as you go along.

Bill Venables, 
CMIS, CSIRO Marine Laboratories, 
PO Box 120, Cleveland, Qld. 4163
Phone:  +61 7 3826 7251  
Fax:    +61 7 3826 7304
Mobile: +61 419 634 642
<mailto: Bill.Venables at csiro.au>

r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list