This track invites presentations in the area of Statistical algorithms and software. Full papers will be considered for publication in the first CSDA special issue which will inaugurate this section. Details can be found below:

The journal "Computational Statistics and Data Analysis" aims to publish, inter alia, papers with a strong algorithmic and software component. These papers will appear in the first section of the journal under the heading - "Computational Statistics".

In addition, the journal offers authors the possibility to enhance their articles by including multimedia files along with the online article. These may be in the form of algorithms or software. These files will appear embedded in the article on ScienceDirect at www.sciencedirect.com/science/journal/01679473.

Over time, an archive of statistical algorithms will be established at this site. The files should be in non-proprietary formats, to ensure that the long-term archival needs of electronic publishing are met. The multimedia files will also be listed on the web page of the "International Association for Statistical Computing".

The files will be freely available at this site. Copyright of the files will remain with the authors and their use will require the citation of the related article (journal, author, title and date of the publication). No liability is assumed by the contributors, the publisher, or the IASC.

Scope:

Significant research contributions relating to the design, analysis and implementation of algorithms and software to support statistical methods and applications will be considered for inclusion in the "Computational Statistics" section of CSDA.

Algorithms and software contributions must be of use in statistical applications. An algorithm will be considered for inclusion if it provides a novel approach to the solution of a statistical problem, or has a performance superior to that of the existing algorithms. This might entail improvements in execution speed, numerical stability, portability, etc.

Submissions:

A software submission must include test data to allow the correctness of the software to be assessed. The test data should be sufficiently comprehensive to demonstrate the robustness of the software in all relevant circumstances.

When presenting the statistical algorithms, the issues of accuracy, complexity and portability must be addressed. A complete program for testing the algorithm should be supplied so that results obtained theoretically can be verified in practice. The algorithm must be complete, portable, well documented, and well structured. There is no standard requirement for pseudo codes. However, implementations should be expressed in a widely used language.

Submissions of brief notes and remarks concerning existing statistical algorithms and software are also encouraged.

Co-Chairs:

James Gentle School of Computational Sciences MS 5C3 George Mason University Fairfax, VA 22030 USA Tel: 703-993-1994 Fax: 703-993-1993 E-mail: jgentle@gmu.edu	John Hinde Department of Mathematics, National University of Ireland, Galway, GALWAY, Ireland Tel: +353 91 492043 Fax: +353 91 494542 E-mail: john.hinde@nuigalway.ie
Cristian Gatu Institut d'informatique, Universite de Neuchatel, Rue Emile-Argand 11, CP2, CH-2007 Neuchatel Switzerland Tel: +41 (0)32 718 2727 Fax: +41 (0)32 718 2701 E-mail: cristian.gatu@unine.ch