Everyone is a speculator by necessity. Frequentists (or I at least) want data on one side, decision-making and therefore risk/reward analysis on the other, separate. This way anyone can impose any loss function they like on the data and its uncertainty estimate(s).
Lots and lots of people get this wrong, of course, even among those who should know better. It's confusing P(A|B) with P(B|A), and Bayes' Theorem relates the two. But frequentists don't actually deny Bayes' Theorem; they just abstain from using it.
-
-
Not quite - frequentists say you can't use probability to represent opinions. The Bayesian probability is an opinion about the world. Bayes rule is changing your opinion when you see new data. Frequentist probability is about the long run occurrence of hypothetical events.
-
The p-val is "what if we ran this exact same experiment a bunch of times and null were true". It's hypothetical because you don't actually do that.
End of conversation
New conversation -
-
-
The converse silliness from the Bayesian side is the "uninformative prior", an oxymoron if there ever were one. Sometimes it's time to admit "we don't have a clue what's going on, but this is the frequency at which X happens, as a function of Y."
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.