This is the mail archive of the gsl-discuss@sources.redhat.com mailing list for the GSL project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Functionality to fit data to statistical distributions


Hello,

I have written some functionality to set up the
'gsl_multifit_fdfsolver' to determine parameters for various
statistical distributions.

Up til now I have only implemented the distributions I have used
needed myself, but if it is of interest to include something like this
in gsl I would be more than happy to complete all the (continuous)
distributions, and improve (i.e. write) the documentation.

The included files are:

 gsl_distfit_core: This sets up some datastructures, and actually
                   calls the gsl solver.
 
 gsl_distfit_eval: This is a long list of small functions which
                   evaluate the various distribution functions, and
                   the derivatives needed for least squares fdf
                   fitting.

In addition there is a Makefile and a small test application:
dist-test.c


Joakim

Attachment: gsl_distfit_core.c
Description: Binary data

Attachment: gsl_distfit_core.h
Description: Binary data

Attachment: gsl_distfit_eval.c
Description: Binary data

Attachment: gsl_distfit_eval.h
Description: Binary data

Attachment: Makefile
Description: Binary data

Attachment: dist-test.c
Description: Binary data



-- 
  /--------------------------------------------------------------------\
 / Joakim Hove  / hove@bccs.no  /  (55 5) 84076       |                 \
 | Unifob AS, Avdeling for Beregningsvitenskap (BCCS) | Stabburveien 18 |
 | CMU                                                | 5231 Paradis    |
 \ Thormøhlensgt.55, 5020 Bergen.                     | 55 91 28 18     /
  \--------------------------------------------------------------------/

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]