An introduction to Jeffreys priors - 1

These series of videos explain what is meant by Jeffreys priors as well as how they satisfy a particular notion of ‘uninformativeness’. This concept is explained through a simple Bernoulli example.
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...

[@nhlanhlawandile6600]

On what distributions can we use the Jeffrreys prior?

[@belarminolopes811]

Normal, Bernoulli ect

[@spyridon6044]

Excellent videos of yours!!! :)

[@conradsnowman]

Ok so, it's just a change of variables as you would do in an integral?

[@rols3376]

so after reading that waffle, I still don't know what a Jefferys prior is.

[@rangjungyeshe]

It's a way of ensuring that the conclusion you reach from updating your original beliefs (summarised by the prior distrib) with your newly-acquired data (summarised by the likelihood) remain the same, regardless of whether the parameter you're interested in is theta (say, a probability of a certain outcome) or a monotonic function of theta (say, the odds, rather than the probability).
If this wasn't the case, you'd have two people reaching different conclusions from the same data, simply because they've chosen "somewhat different" (aka monotonically related) ways to capture what they're interested in - which would be a bit weird.
Indeed, R A FIsher, one of the giants of 20th century statistics, regarded this potential problem in Bayes to be fatal, and he went on to develop "frequentist" methods, which simply duck the problem - and cause all sorts of far worse problems instead!

[@limzijian98]

@@rangjungyeshe so basically, its something similar to irregardless of what kind of units you are using , you will still get the same answer ?