[SciPy-User] Wording question regarding to distributions

[josef.pktd at gmail.com]

On Thu, Jan 14, 2010 at 5:37 PM, David Warde-Farley <dwf at cs.toronto.edu> wrote:
>
> On 14-Jan-10, at 5:30 PM, Gökhan Sever wrote:
>
>> Hello,
>>
>> What is the right way to express:
>>
>> Do we fit data to a distribution or distribution to data?
>>
>> Thanks.
>
> I would say the latter. Assuming we are talking about the same
> scenario, the data follow some unknown distribution, which you try to
> approximate with some parametric form using maximum likelihood
> estimators and such. So you are fitting a (particular) distribution
> (or more specifically, a model of the underlying process which *uses*
> that particular distribution) to observed data.

I agree, unless you are "massaging" your data to fit the distribution
to get nicer results.

Josef

>
> My $0.02,
>
> David
> _______________________________________________
> SciPy-User mailing list
> SciPy-User at scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-user
>