Would like to develop statistical subroutines

[John Graham-9]

Hi there,

I want to start contributing statistical subroutines for PSPP - I was just wondering if there are any that are much requested/desired so that I can get on with those?

I've read all the info on the 'contributing to PSPP' page of the website, and was planning to just start working through the list of things given there (i.e. "Logistic regression, Poisson regression, smoothing splines"... etc.) but if there's something that's desperately wanted I'll make a start on that instead.  Alternatively, is there (or could there be?) a list of functions needed, so people wanting to contribute will know where to make best use of their time?

Cheers,

John G

_______________________________________________
pspp-dev mailing list
pspp-dev@...
http://lists.gnu.org/mailman/listinfo/pspp-dev

[Ben Pfaff]

"John Graham" <johngavingraham@...> writes:

> I want to start contributing statistical subroutines for PSPP - I was just
> wondering if there are any that are much requested/desired so that I can get
> on with those?

That sounds great.

A good place to start looking for features to implement is this:
go to sv.gnu.org/p/pspp, click on Bugs, switch to the Advanced
query form under Display Criteria (and click on Apply), then
click on the Severity column, twice (to sort in ascending order
of Severity).  Then all the Wishlist bugs should be at the top,
and you can quickly see what statistical features have been
requested.
--
"Because computer source code is an expressive means for the exchange
 of information and ideas about computer programming, we hold that it
 is protected by the First Amendment."
--Hon. Boyce F. Martin, Jr., for the 6th Circuit Court, Junger vs. Daley


_______________________________________________
pspp-dev mailing list
pspp-dev@...
http://lists.gnu.org/mailman/listinfo/pspp-dev

[John Darrington]

A list of unimplemented commands can be found here:

 http://www.gnu.org/software/pspp/manual/html_node/Not-Implemented.html

It might be a good idea if someone were to copy these to
http://jstover.motd.org in such a way that people can "vote" for the
procedures that they would most want to see.


My impressions from talking to users, and browsing the users list and
other forums, is that PSPP's most urgently required statistical
procedures are:

 1. Multivariate and repeated measures anova.  I understand Jason is
    working on a generalised linear model procedure, so all the others
    like MANOVA, ANACOVA, etc will be straightforward once the
    groundwork has been done.

 2. Logistic Regression.

 3. Reliability analysis.  I'm working on a cut-down version of
    RELIABILITY ; it'll do cronbach's alpha and a few other things,
    which is what most people use it for so far as I can tell.

 4. Factor Analysis.  Some reading has led me to conclude the most
    tricky bit will be aspects such as varimax rotation.

 5. Non-parametric methods, such as Kolmogorov-Smirnov, Wilcoxon, and
    Mann-Whitney tests.

J'






On Thu, Sep 04, 2008 at 02:00:04PM +0100, John Graham wrote:
     Hi there,
     
     I want to start contributing statistical subroutines for PSPP - I was just
     wondering if there are any that are much requested/desired so that I can get
     on with those?
     
     I've read all the info on the 'contributing to PSPP' page of the website,
     and was planning to just start working through the list of things given
     there (i.e. "Logistic regression, Poisson regression, smoothing splines"...
     etc.) but if there's something that's desperately wanted I'll make a start
     on that instead.  Alternatively, is there (or could there be?) a list of
     functions needed, so people wanting to contribute will know where to make
     best use of their time?
     
     Cheers,
     
     John G

     _______________________________________________
     pspp-dev mailing list
     pspp-dev@...
     http://lists.gnu.org/mailman/listinfo/pspp-dev


--
PGP Public key ID: 1024D/2DE827B3
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://pgp.mit.edu or any PGP keyserver for public key.




_______________________________________________
pspp-dev mailing list
pspp-dev@...
http://lists.gnu.org/mailman/listinfo/pspp-dev