[SciPy-User] [OT] Bayesian vs. frequentist

josef.pktd@gmai... josef.pktd@gmai...
Fri Feb 17 21:21:06 CST 2012


On Tue, Feb 14, 2012 at 3:39 PM, Sturla Molden <sturla@molden.no> wrote:
> On 14.02.2012 21:24, josef.pktd@gmail.com wrote:
>
>> Do you expect an argument? sounds a bit like http://andrewgelman.com/
>
> No I don't. I am just tired of explaining that "significant p-value"
> does not imply "very important effect". Sorry for spamming the list with
> my rant.

Since I just came across a paper on the topic
http://www.anesthesia-analgesia.org/content/112/3/678.short while
looking for something else.

I have seen now several papers (mainly the abstracts) that test
"equivalence" instead of "equality", so we can specify in advance what
difference is "important" and what would be "equivalent".

(not Bayesian)

Josef

>
> Sturla
>
>
>
>
>
> _______________________________________________
> SciPy-User mailing list
> SciPy-User@scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-user


More information about the SciPy-User mailing list