[Numpy-discussion] statistical model fitting & comparison

Bruce Southey bsouthey@gmail....
Tue Oct 21 16:01:10 CDT 2008


Stéfan van der Walt wrote:
> Hi Joris
>
> 2008/10/21 Joris De Ridder <Joris.DeRidder@ster.kuleuven.be>:
>   
>> I'm interested in developing some general-use Python/Numpy code for
>> linear model fitting and comparison. The fitting is easy enough with
>> Numpy, but the automated comparison of the submodels to identify which
>> model describes best the data, requires some work. Before I embark on
>> this, I would like to find out if anyone actually already wrote
>> something along these lines, in order not to reinvent the wheel.
>>     
>
> I would guess that this is part of Jonathan Taylor's "models" package,
> but I may be wrong.  It is available at the Neuroimaging in Python
> homepage:
>
> http://neuroimaging.scipy.org/
>
> You may also want to look at the optimisation scikit and the RANSAC
> cookbook example.
>
> Regards
> Stéfan
> _______________________________________________
> Numpy-discussion mailing list
> Numpy-discussion@scipy.org
> http://projects.scipy.org/mailman/listinfo/numpy-discussion
>
>   
Hi,
I think you are on your own here as it is a huge chunk to chew! 
Depending on what you really mean by linear models is also part of that 
(the Wikipedia entry is amusing). Most people probably to stats 
applications like lm in R and glm in SAS.

I do have an interest and the code I have (for logistic regression) is 
probably just as easy to rewrite.

Bruce



More information about the Numpy-discussion mailing list