RISC-Linz RISC-Linz Research Institute for Symbolic Computation  
about
|
people
|
publications
|
research
|
education
|
industry
|
conferences
|
media
|
projects
internal
description  |  members  |  seminar  |  publications  |  software  |  events  |  links
  
search:
  

The Paule/Schorn Implementation of Gosper's and Zeilberger's Algorithms

Short Description

With Gosper's algorithm you can find closed forms for indefinite hypergeometric sums. If you do not succeed, then you may use Zeilberger's algorithm to come up with a recurrence relation for that sum. Both algorithms may be used to find and prove identities involving hypergeometric terms and sums of those.

Registration and Legal Notices

The source code for this package is password protected. To get the password send an email to Peter Paule. It will be given for free to all researchers and non-commercial users.

Copyright © 1999–2011 The RISC Combinatorics Group, Austria — all rights reserved. Commercial use of the software is prohibited without prior written permission.

A Note on Encoded Files

This package contains one or more Mathematica input files which are encoded. Those files cannot be read or modified directly as plain text, but can be loaded into Mathematica just like any normal input file (i.e., with <<"file" or Get["file"]). There is no need (and also no way) to decode them by using additional software or a special key.

If loading an encoded file causes a syntax error, open it with a text editor and remove any blank lines at the beginning (for some reason your Mac could have inserted them silently...).

The Package

The Paule/Schorn implementation of Gosper's and Zeilberger's algorithm is contained in the Mathematica input file and is accompanied by the Mathematica notebook

Screenshot

Click here to see Gosper's and Zeilberger's algorithms in action.

Literature

To use the implementation it is sufficient to study the notebook readme.nb. It contains a few examples to start with.

You find some easy and several more involved examples for using the implementation in our joint paper

P. Paule and M. Schorn, A Mathematica Version of Zeilberger's Algorithm for Proving Binomial Coefficient Identities, J. Symbolic Comput., 20 (1995), 673-698. [pdf]
The diploma thesis of Markus Schorn makes sure that the algorithm is indeed correct. Our implementation is closely related to the detailed background developed there. And again it contains several interesting examples.
M. Schorn, Contributions to Symbolic Summation, Diploma Thesis, RISC, J. Kepler University, Linz, December 1995. [pdf]
Methods for tuning the algorithm are described in the paper
A. Riese, Fine-Tuning Zeilberger's Algorithm: The Methods of Automatic Filtering and Creative Substituting, in Symbolic Computation, Number Theory, Special Functions, Physics and Combinatorics (F.G. Garvan and M.E.H. Ismail, eds.), Developments in Mathematics, Vol. 4, pp. 243-254, Kluwer, 2001. [pdf]
Finally, in the book A = B by M. Petkovsek, H. Wilf, and D. Zeilberger you find a collection of methods to automatically prove identities.

Versions and Bugs

Right now you are using Version 3.52 last updated on January 12, 2005. Please report any bugs and comments to Peter Paule or Ralf Hemmecke.