I have to write a math history paper. I was going to write it on the rise of Bayes' Theory. I have read around that Bayes' theory was no widely accepted or used until the 20th century. I need to make a claim and have a view point as to why this was the case for my thesis but I am having trouble doing so. Does anyone have any ideas?
$\begingroup$
$\endgroup$
6
-
1$\begingroup$ Coincides with rise in computational power. $\endgroup$– user2974951Commented Apr 8, 2021 at 6:10
-
$\begingroup$ I don't know for sure, but my impression from ET Jaynes' papers is that it actually was used and accepted from its invention up until around the early-mid 20th century when frequentism arose and became the mainstream view. Before that nobody was really a "Bayesian" or a "frequentist", they just used whatever worked. $\endgroup$– N. VirgoCommented Apr 8, 2021 at 10:25
-
$\begingroup$ A good start would be to review relevant sections of Todhunter's history (1865), with a focus on Bayes and Laplace. $\endgroup$– whuber ♦Commented Apr 8, 2021 at 13:15
-
$\begingroup$ Can you add references to what you ave read? $\endgroup$– kjetil b halvorsen ♦Commented Apr 8, 2021 at 15:00
-
$\begingroup$ There are explanations on different levels. Here are two further details, in practice more important than might seem right. First, the word subjective was fated to divide people into opposing camps. Second, personalities. Several prominent Bayesians were idiosyncratic, quirky, non-joiners, sometimes delighting in controversy and even polemics. As such they were less likely to have influential students or collaborators. This seems to have applied in varying degrees to H. Jeffreys, B. De Finetti, L.J. Savage, I.J. Good, E.T. Jaynes and even D.V. Lindley from the 1930s for some while. $\endgroup$– Nick CoxCommented Apr 8, 2021 at 16:19
|
Show 1 more comment
1 Answer
$\begingroup$
$\endgroup$
1
"Versions of the Bayesian approach were applied to scientific work in the late 1700s and repeatedly in the 19th century. But after World War I, anti-Bayesian statisticians, like Sir Ronald Fisher, succeeded in marginalizing the approach. All Fisher said about Bayesian analysis (then called inverse probability) in his influential 1925 handbook was:
[...] the theory of inverse probability is founded upon an error, and must be wholly rejected.
Bayesian data analysis became increasingly accepted within statistics during the second half of the 20th century, because it proved not to be founded upon an error. All philosophy aside, it worked. Beginning in the 1990s, new computational approaches led to a rapid rise in application of Bayesian methods."
From the book statistical rethinking by Richard McElreath. He provide reference for further info. in the book, here is the link to the book:
-
$\begingroup$ McElreath is stimulating but (I guess) be the first to admit his partiality. In defence of Fisher, I would say that his advocacy of likelihood places him nearer Bayesian views than this description implies, not that he would readily have admitted it. But he doesn't deserve all the credit. 19th century writers on probability like John Venn created the climate in which Fisher was trained $\endgroup$– Nick CoxCommented Apr 8, 2021 at 16:23