SAMSI Announces Upcoming Research ProgramsNovember 20, 2005
The Statistical and Applied Mathematical Sciences Institute (SAMSI) recently entered its fourth year of operation. Located in Research Triangle Park, North Carolina, SAMSI is a National Science Foundation-supported institute focused on the synthesis of the statistical sciences with the applied mathematical sciences and disciplinary science to confront the very hardest and most important data- and model-driven scientific challenges.
After a very successful third year, with well over 700 participants in its activities, SAMSI is running three programs in the academic year 2005�06: Financial Mathematics, Statistics and Econometrics; National Defense and Homeland Security; and Astrostatistics. Opportunities for participation in these programs still remain, especially for the program in astrostatistics, which begins in January 2006 and has activities and workshops running through July 2006. Readers can visit the SAMSI Web site (www.samsi.info) for further information about these programs.
Plans are well under way for SAMSI's 2006�07 programs, which offer numerous opportunities for participation by members of the SIAM community. New and senior researchers can visit SAMSI for periods of one month to one year; new researchers benefit both from the SAMSI environment and from financial support, and senior researchers have the opportunity to broaden their interests and skill sets.
Several postdoctoral positions are available for each SAMSI program. Special programs have been designed for graduate and upper-level undergraduate students so that they can begin their involvement in cross-disciplinary and team research.
Each SAMSI program also has at least an opening and a closing workshop, making it possible for indidviduals who cannot spend part of the year at SAMSI to participate in these research efforts.
New researchers and members of underrepresented groups are especially encouraged to participate in SAMSI workshops and programs.
SAMSI programs for 2006�07 are:
1. High Dimensional Inference and Random Matrices, September 2006 through December 2006. Massive data collection efforts across contemporary science, technology, and commerce are yielding datasets of ever higher dimensions. Often, the dimensionality (number of variables p) is comparable to, or even larger than, the number of cases (sample size n), and there is a crucial need for fundamental tools and theory that can provide reliable understanding of such problems. At the same time, recent advances within random matrix theory offer new mathematical tools and approaches that can be brought to bear on the spectral aspects of high dimensional data. Under the direction of Iain Johnstone, Helene Massam, and Craig Tracy, this program will bring together statisticians, applied mathematicians, and probabilists to study these problems in the context of methodologically intensive application domains.
2. Development, Assessment, and Utilization of Complex Computer Models, September 2006 through May 2007 (to be preceded by a summer school and other activities in the summer of 2006). Mathematical models intended for computational simulation of complex real-world processes are a crucial ingredient in virtually every field of science, engineering, medicine, and business. A variety of common challenges arise in the development, evaluation, and use of complex computer models, including integration of experimental data, simultaneously coping with the many varieties of uncertainty that result, validation and improvement of the model, approximation of the model over a possibly high-dimensional input space, and optimization and design of computer experiments themselves. Participants in this program will study these and other issues in the context of subprograms involving some subset of engineering models, biomedical models, environmental/ecological models, social network models, microsimulation models, and dynamical models.
In addition to these two major programs, SAMSI will conduct planning or hot-topics workshops, undergraduate and graduate outreach workshops, summer schools (as part of programs), and several interdisciplinary courses associated with ongoing programs. A new initiative at SAMSI is to conduct intensive summer programs--which will typically run for two weeks--in more methodological or theoretical areas of statistics and applied mathematics. The first such program, on multiple testing, is scheduled for late July 2006. This program will investigate the readily apparent discrepancy in practice between nominal error probabilities in hypothesis testing and the actual proportion of true null hypotheses, while exploring the options available for adjustment in the presence of multiple hypothesis tests. This intensive summer program will be led by Peter Mueller, Juliet Schaffer, Stan Young, and Peter Westfall. Ideas for other summer programs are welcomed by the SAMSI directorate.
With a proposal for renewed NSF support now in preparation, SAMSI encourages readers to propose topics for future research programs. Suggestions can be directed to SAMSI's director, James Berger ([email protected]);
to the other members of the SAMSI directorate: Chris Jones ([email protected]), Alan Karr ([email protected]), and Ralph Smith ([email protected]); or the SAMSI National Advisory Committee, chaired by Peter Bickel and Carlos Castillo Chavez.
Additional information about SAMSI activities and opportunities for participation can be found at the SAMSI Web site: www.samsi.info.