[SciPy-User] scipy.stats.fit inquiry

josef.pktd@gmai... josef.pktd@gmai...
Tue Oct 20 11:50:37 CDT 2009


On Tue, Oct 20, 2009 at 11:13 AM, Anne Archibald
<peridot.faceted@gmail.com> wrote:
> 2009/10/20  <josef.pktd@gmail.com>:
>
>> I never heard about the Cash statistic.
>
> It's a clever trick for estimating uncertainties on fitted parameters;
> you do some magic with the likelihood ratio and you get statistic that
> behaves like chi-squared, apart from being exactly zero at your
> best-fit value. So it's no use for esstimating quality-of-fit, but you
> can use it to get error regions just the way you would if you'd had
> Gaussian statistics and a chi-squared fit. (Cash 1979, "Parameter
> estimation in astronomy through application of the likelihood ratio")
>
> Incidentally, I have some code implementing the Kuiper test, a
> modified K-S test that is sensitive to different aspects of the shape
> of the distribution, and (more importantly for me) is invariant on
> shifting a distribution or sample modulo 1. I haven't submitted it for
> inclusion because the interface I used is a little different from that
> used by scipy's K-S test, but if there's interest I'd be happy to
> contribute it.

More tests in scipy.stats is always good (at least as long as I don't
have to chase down the references to write the tests for the tests.)
How do you get the pvalue or critical values for Kuiper, since the
distribution of the test statistic is not a standard distribution?

(From what I understand from Sherpa, is that Cash is used as
objective function for the maximum likelihood estimation of a
Poisson process.)

Looking for Kuiper, I found a nice overview of a large list of goodness
of fit tests, used in natural sciences.

http://www.ge.infn.it/statisticaltoolkit/gof/deployment/userdoc/statistics/applicability.html

with article in
http://ieeexplore.ieee.org/xpls/abs_all.jsp?isnumber=29603&arnumber=1344284&count=103&index=22


And apropos circular statistic, which I know nothing about:

stats.distribution.vonmises is the only distribution that has bounded
(or circular) support but doesn't define the bounds (a, b). This
screws up numerical integration, e.g. for the moment calculation.

Is vonmises actually used on circular support?
To enable integration, we need to define a and b (e.g [-pi,pi] as
in numpy random) or introduce new bounds specifically for
integration in this case.

Josef

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


More information about the SciPy-User mailing list