The illusion of uninformative priors
Вставка
- Опубліковано 6 вер 2024
- This video explains why priors always - in some frame of reference - convey information and, hence, can not, as might be expected, be devoid of information.
This video is part of a lecture course which closely follows the material covered in the book, "A Student's Guide to Bayesian Statistics", published by Sage, which is available to order on Amazon here: www.amazon.co....
For more information on all things Bayesian, have a look at: ben-lambert.co.... The playlist for the lecture course is here: • A Student's Guide to B...
This is a compelling argument, and I agree: there really is no way of choosing the "right" prior distribution based on no data whatsoever, and even if you could that prior would not be consistent.
How did we just remove the integrals at 7:40?
Because they aren’t necessary, the point is that the area under the curve and the pdf itself are equivalent, he just put them there at first for clarity
The title is quite misleading, since you talk about the principle of indifference. You can define 'uninformative' otherwise, and then the concept becomes meaningful.
Nice example - not seen it before.
Also at 8:11, why is P(theta) just 1?
Because it is a uniform distribution and has value 1 everywhere?
Yes, p(theta) is proportional to 1 because it’s uniform