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

denis denis-bz-gg@t-online...
Fri Feb 17 04:46:27 CST 2012


Sturla,
  that's funny.
In the same vein, there are two kinds of customers:
- smart, but no money
- money, but no smarts.

On a related split see Breiman, "Statistical Modeling: The Two
Cultures"
http://projecteuclid.org/euclid.ss/1009213726 2001 17p + 16 debate
related because many splits are cultural, e.g. practitioners vs
academics.
I can't begin to summarize Breiman -- read it, it's really good.
See also
http://stats.stackexchange.com/questions/6/the-two-cultures-statistics-vs-machine-learning

cheers
  -- denis

On Feb 14, 8:40 pm, Sturla Molden <stu...@molden.no> wrote:
> After having worked with applied statistics for ~15 years, I have
> reached this conclusion... ;-)
>
> Sturla's 20 propositions on Bayesian vs. classical statistics:



More information about the SciPy-User mailing list