All entries for Sunday 02 October 2016

October 02, 2016

Classical statistics revisited


I’ve written before about the use of the term “classical” to refer to traditional frequentist statistics. I recently found that E.T Jaynes had covered this ground over 30 years ago. In “The Intuitive Inadequacy of Classical Statistics” [1] he writes:

“What variety of statistics is meant by classical? J.R. Oppenheimer held that in science the word “classical” has a special meaning: “[…] it means “wrong”. That is, the classical theory is the one which is wrong, but which was held yesterday to be right.”

“… in other fields, “classical” carries the opposite connotations of “having great and timeless merit." Classical music, sculpture and architecture are the kind I like.”

Jaynes follows convention, and Oppenheimer, in the article and means traditional stats by “classical”. I guess the Oppenheimer meaning should be understood more generally.

[1] Epistemologia VII (1984) Special Issue. Probability, Statistics and Inductive Logic pp 43-74

October 2016

Mo Tu We Th Fr Sa Su
Sep |  Today  | Nov
               1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30

Search this blog



Most recent comments

  • Hi Tom Sorry for delay in replying – taken out by family issues then holiday for the last month or s… by Simon Gates on this entry
  • Simon, I can see where you're coming from on this. If MCID (in its various guises) is not an optimal… by Chee-Wee Tan on this entry
  • Hi Simon I am currently doing my PhD in clinical based research. We want to use the MCID to determin… by tomwilks on this entry
  • I think your comment reveals how nonsensical null hypothesis testing is (and I see from your other p… by matt on this entry
  • Thanks for commenting Matt – I do wonder if anyone ever looks at any of this, not that this is a pro… by Simon Gates on this entry

Blog archive

Not signed in
Sign in

Powered by BlogBuilder