David A. Heiser wrote:
> First of all Fisher is a very ponderous writer, very difficult to find the
> gold in the pile of ore.
> Second, one needs to read Fisher's insight into
> Bayes original work to understand Fisher's view of probability.
Can you supply a reference? All Fisher's writings on `Bayesian'
ideas that I have read just give me the conviction that he never
even began to understand what was being suggested by non-
frequentists.
If you are meaning `fiducal inference' I've never managed to grasp
the nub of what was being proposed; and would welcome references
[IIRC some of the big names of his era never believed that he was
proposing a generally applicable mechanism.]
> I suspect
> that von Mises was frustrated with Fisher's English, which could not convey
> the subtleties of meaning that can be conveyed by technical German.
Not wishing to start linguistic wars: I suspect that any weaknesses
of Fisher's english was just that -- a weakness in his use of the
language rather than hitting the boundaries of its expressiveness.
[Of course effective use of language and anticipation of readers
problems are strongly linked.]
I remember an anecdote where a student went up to Fisher after a
lecture asking him to explain a how an expression was derived (there
was a rather large gap in a proof). Fisher thought about it for several
minutes and said something like `but it's obvious' -- and was apparently
unable (rather than unwilling) to explain further.
Peter
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
