The workshop was organized jointly with the 5th International Workshop on Parallel Matrix Algorithms and Applications (PMAA'08) and the 2nd International Conference on Computational and Financial Econometrics. All three workshops were founded within the framework of the ERCIM Working Group and were organized by its members. PMAA'08, which was sponsored by the ERCIM WG, was organized in honour of Bernard Philippe (IRISA, Rennes, France), cofounder of the previous WG on Matrix Computations and Statistics. PMAA'08 had four plenary talks, more than eighty sessions with some 380 presentations, and over 400 participants.
Plenary talks were given by Bernard Philippe, INRIA-IRISA, Rennes, France on "A parallel GMRES method preconditioned by a Multiplicative Schwarz iteration. Michael W. Berry fom the University of Tennessee, Knoxville, USA, gave a presentation entitled: "Exploiting nonnegativity in matrix and tensor factorizations to improve topic detection and tracking in text mining". Oliver Linton, London School of Economics and Political Science, UK gave "The FINRISK (Swiss National Centre of Competence in Research "Financial Valuation and Risk Management") Lecture: Iterative smoothing algorithms and their application in finance"; and Herman Van Dijk, Erasmus University Rotterdam, The Netherlands presented "The CSDA (Computational Statistics and Data Analysis) Lecture: Possibly ill-behaved posteriors in econometric models: On the connection between model structures, non-elliptical credible sets and neural network Simulation"
The workshop also provided an occasion to discuss the activities of the Working Group, enhance network activities among its members, and consider future research collaboration within European projects. The next ERCIM Working Group meeting will take place in Cyprus in October 2009.
The ERCIM Working Group "Computing and Statistics focuses on all computational aspects of statistics. Of particular interest is research in important statistical applications areas where both computing techniques and numerical methods have a major impact. All aspects of statistics which make use, directly or indirectly, of computing are considered. Applications of computational statistics in diverse disciplines will be strongly represented. These areas include economics, medicine and epidemiology, biology, finance, physics, chemistry, climatology and communication. The aimof the WG is twofold: first, to consolidate the research in computational statistics that is scattered throughout Europe; second to provide researches with a network from which they can obtain an unrivalled source of information about the most recent developments in computational statistics and applications.
ERCIM Working Groups are open to any researcher in the specific scientific field. Scientists interested in participation should contact the WG coordinator.
ERCIM Computing and Statistics WG coordinator, University of London