[SciPy-user] [SciPy-dev] Statistics review months

Alan Jackson alan at ajackson.org
Sun Apr 2 13:51:45 CDT 2006


On Sat, 01 Apr 2006 20:20:02 -0800
Jonathan Taylor <jonathan.taylor at stanford.edu> wrote:

> on this topic, as an honest-to-goodness statistician it might be nice to 
> see more statistical modelling in scipy. i know Rpy exists, but the 
> interface is not very pythonic.
> 
> i have some "home-brew" modules for linear regression, formula building 
> (something like R's) and a few other things. if it went into something 
> like scipy, it might gain from the criticisms of others....
> 
> is there any interest in making the equivalent of a
> 
> scipy.stats.models
> 
> module?
> 
> i think an easily (medium-term) achievable goal is:
> 
> i) linear (least-squares) regression models with/without weights or 
> non-diagonal covariance matrices (in R: lm + more)
> 
> ii) generalized linear models (in R: glm)
> 
> iii) iteratively reweighted least squares algorithms (glm is a special 
> case), i.e. robust regression  (in R: rlm).

I'm a big fan of R and of rlm in particular. I have to agree with your
comments about Rpy, though I think their plans for it head in the right
direction and give hope of a better interface.
But, yes, I would support adding those capabilities to SciPy. I have Rpy
accessing rlm in a little product right now, and it would certainly
simply life!

> 

-- 
-----------------------------------------------------------------------
| Alan K. Jackson            | To see a World in a Grain of Sand      |
| alan at ajackson.org          | And a Heaven in a Wild Flower,         |
| www.ajackson.org           | Hold Infinity in the palm of your hand |
| Houston, Texas             | And Eternity in an hour. - Blake       |
-----------------------------------------------------------------------



More information about the SciPy-user mailing list