------------------------------------------------------------- This mail is going to both the participants of the RSR workshop in Banff and the r-sig-robust mailing list; so there will be some cross-posting; sorry for this, I did not know how to avoid this. ------------------------------------------------------------- As announced in the (end-)minutes of the Banff workshop, I have now set up an R-forge project "robust-ts". see: https://r-forge.r-project.org/ and https://r-forge.r-project.org/projects/robust-ts/ This project is meant as a platform for collaborative implementation of robust counter-parts to the time series functionality in R-package "stats". On the long run ---be it by direct integration of S-Plus code if legal (Kjell Konis is pushing this forward at Insightful) or by re-implementation--- it is also to contain the robust time series routines of chapter 8 of "Robust Statistics, Theory and Methods" by Maronna, Martin and Yohai; 2006. So far, not too much has been done (in particular: there is not a single bit of code up to now) and some things have still to be settled with the Vienna guys who provide the R-forge infrastructure. In order not to blow up this mail unnecessarily, I have written up two small .txt-Files The first one lists some open issues in using R-forge for this project, which remain to be settled in the next future. The second is a step-by-step "Howto" for anyone who wants to collaborate in this project. Any volunteer in this direction will be warmly welcome! These files are available under https://r-forge.r-project.org/plugins/scmsvn/viewcvs.php/www/OpenIssues.txt?rev=3&root=robust-ts&view=markup https://r-forge.r-project.org/plugins/scmsvn/viewcvs.php/www/HOWTO-collaborate.txt?rev=3&root=robust-ts&view=markup Anyway, the first step is taken Best, Peter
[RsR] R-forge project "robust-ts" (the time series counter-part to "robustbase") is set up
1 message · Peter Ruckdeschel