Volume 40 • Issue 6 IMS Bulletin
September 2011 IMS Annual Meeting at JSM
Contents The Institute of Mathematical Statistics held its 74th annual meeting at the Joint 1 IMS Annual Meeting at JSM Statistical Meetings in Miami’s South Beach, from July 30 to August 4, 2011. The IMS highlights at JSM included the three Wald lectures by Greg Lawler, the Neyman 2 Members’ News: Christian Genest, CR Rao, Joe Glaz, lecture by Michael I. Jordan, and five Medallion lectures, by Jianqing Fan, Chris Xuming He, Jean-Pierre Holmes, Michael Newton, Sylvia Richardson, and Qi-Man Shao (see photos on page Fouque, Craig A. Tracy 5). The IMS awards session included the presentation of IMS Fellows (photo 3 COPSS Awards on page 3), the Carver Award (to Ross ommons/Alexf
Project Euclid news; C 4 Leadbetter—page 5) and the Laha Travel Web-mentoring project Award recipients (page 7). It was followed W 5 IMS@JSM in photos by Peter Hall’s Presidential Address,
6 Anirban’s Angle: Berkeley Photo: ikimedia which is reproduced on pages 12–14 of and Indian statistics this issue. Miami in August is hot and very 7 Laha awardees humid, but despite that, many attendees 8–9 Obituaries: AV Skorokhod, said that they liked the Art Deco district, The Art Deco Marlin Hotel in South Beach Patrick Billingsley the Cuban restaurants, and of course, 10 Rick’s Ramblings: the beach. With temperatures outside in the mid-90s (about 34ºC) and humidity at NSF proposal-writing around 85%, the 5,000 participants were certainly grateful for the air-conditioning! 12 Presidential Address: Peter At the Business Meeting, outgoing President Peter Hall summarized IMS activities Hall over the previous year, and paid tribute to all those who have worked on IMS com- mittees. He thanked former Past-president Mike Steele, who 15 SPA meeting report has now finished his term on the Executive Committee, for 16 BJPS 25th anniversary his role in shepherding 17 Terence’s Stuff: Speaking, the institute’s financial reading, writing resources. Peter passed the 18 IMS meetings gavel to the new 20 Other meetings President, Ruth 21 Employment Opportunities Williams. Ruth will introduce 24 International Calendar of Statistical Events herself to IMS Bulletin readers 27 Information for Advertisers in the next issue, and will describe New IMS President a few of the IMS Ruth Williams (left) activities that are receives the gavel from Peter Hall, marking planned or already the transition of the underway. presidency for 2011 IMS Bulletin 2 . IMS Bulletin Volume 40 . Issue 6 Volume 40 • Issue 6 September 2011 IMS Members’ News ISSN 1544-1881 Statistical Society of Canada 2011 Gold Medal awarded to Christian Genest Contact information Christian Genest, Professor of Statistics at McGill University, Montréal (Canada), is the IMS Bulletin Editor: Dimitris Politis recipient of the Statistical Society of Canada (SSC) 2011 Gold Assistant Editor: Tati Howell Medal. This award, which is the Society’s highest distinction, was Contributing Editors: given to Christian “in recognition of his remarkable contributions Peter Bickel, Anirban DasGupta, Rick Durrett, Nicole Lazar, Terry Speed to multivariate analysis and nonparametric statistics, notably through the development of models and methods of inference for Contact the IMS Bulletin by email: studying stochastic dependence, synthesizing expert judgments and multi-criteria decision making, as well as for his applications e [email protected] thereof in various fields such as insurance, finance, and hydrology.” The award ceremony took place at the SSC 39th Annual To contact the IMS regarding your dues, Christian Genest membership, subscriptions, orders or Meeting, held in Wolfville (Nova Scotia), June 12-15, 2011. change of address: For additional details, see http://www.ssc.ca/en/award-winners/award-winners-2011 IMS Dues and Subscriptions Office 9650 Rockville Pike, Suite L3503A Bethesda, MD 20814-3998 CR Rao awarded 33rd honorary degree USA C.R. Rao, Sc.D. (Cantab), FRS, has received another honorary degree, Doctor of Science t 877-557-4674 [toll-free in USA] (Honoris Causa), this time from the University of Colombo, Sri Lanka. It was conferred by t +1 216 295 5661[international] Prof. T. R. Ariyaratne, Dean of the Faculty of Science, who said of Prof. Rao, “He is among f +1 301 634 7099 the world leaders in statistical science over the last six decades. His research, scholarship, e [email protected] and professional services have had a profound influence on theory and applications of statistics.” Professor Rao has now received 33 Honorary Doctoral Degrees from universi- To contact the IMS regarding any other matter, including advertising, copyright ties in 18 countries spanning six continents. Among his recent awards was the UK Royal permission, offprint orders, copyright Statistical Society’s Guy Medal in Gold, awarded triennially to those judged to have mer- transfer, societal matters, meetings, fellows ited a signal mark of distinction by reason of their innovative contributions to the theory nominations and content of publications: or application of statistics. Executive Director, Elyse Gustafson IMS Business Office On the Move… SIAM Fellows PO Box 22718, Beachwood Joe Glaz appointed dept head at U Conn OH 44122, USA SIAM, the Society for Industrial and t 877-557-4674 [toll-free in USA] Joseph Glaz, Fellow of IMS, is now Head Applied Mathematics, has elected 34 t +1 216 295 5661[international] of Department of Statistics, University of new Fellows, among whom are two IMS f +1 216 295 5661 Connecticut, effective July 1, 2011. members, Jean-Pierre Fouque, University e [email protected] of California at Santa Barbara, and Craig Xuming He moves to Michigan A. Tracy, University of California at Davis. Former IMS Bulletin Editor Xuming He has Their citations are, respectively: Executive Committee taken the position of H.C. Carver Professor Jean-Pierre Fouque: For contributions President: Ruth Williams of Statistics, Department of Statistics, to asymptotic analysis for random media and [email protected] University of Michigan, after 18 years financial mathematics. President-Elect: Hans R. Künsch [email protected] of service at the University of Illinois at Craig A. Tracy: For fundamental contri- Past President: Peter Hall Urbana-Champaign (1993–2011). butions to statistical physics, integrable systems, [email protected] and probability Treasurer: Jean Opsomer theory including ran- [email protected] Changing job? Moving house? Don’t forget to tell the IMS! dom matrix theory Program Secretary: Guenther Walther [email protected] Email the Dues & Subs Office and its applications. Executive Secretary: Aurore Delaigle at [email protected] [email protected] IMS Editors September . 2011 IMS Bulletin . 3 IMS Journals and Publications Annals of Statistics: Peter Bühlmann and Tony Cai http://imstat.org/aos Annals of Applied Statistics: Bradley Efron, Karen Kafadar, COPSS Awards Susan Paddock, Tilmann Gneiting, Samuel Kou, Kenneth Lange & Stephen Fienberg http://imstat.org/aoas The Committee of Presidents of Statistical Societies announced its awards, presented at its Annals of Probability: Ofer Zeitouni annual ceremony at JSM in Miami Beach on August 3 by COPSS president Xihong Lin. http://imstat.org/aop The winner of the President’s Award is Nilanjan Chatterjee of the National Cancer Annals of Applied Probability: Andrew Barbour Institute, USA: “For outstanding contributions to the statistical sciences by ingenious method- http://imstat.org/aap Statistical Science: Jon Wellner ological research with applications in epidemiology and genetics, including studies of gene-envi- http://imstat.org/sts ronment interactions, disease heterogeneity and genome-wide association studies; for fundamen- IMS Lecture Notes – Monograph Series tal contributions to the theory of case-control studies and complex retrospective sampling designs; http://imstat.org/publications/lecnotes.htm IMS Collections for demonstrating leadership and a vision as a statistical scientist by actively collaborating in http://imstat.org/publications/ wide-ranging studies of cancer epidemiology and genetics and concurrently maintaining a vigor- imscollections.htm ous methodological research program closely tied to cutting edge scientific issues; for exceptional NSF-CBMS Regional Conference Series in Probability and Statistics: mentoring and service to the profession and to the National Cancer Institute.” Unusually, http://imstat.org/publications/nsf.htm Nilanjan was also selected to receive the Snedecor award, for his contribution to biometric IMS Co-sponsored Journals and research and also recognized for the 2009 JASA paper, co-authored with Y.H. Chen and Publications R.J. Carroll, “Shrinkage Estimators for Robust and Efficient Inference in Haplotype-Based Electronic Journal of Statistics: David Ruppert Case-Control Studies.” http://imstat.org/ejs Electronic Journal of Probability: Bálint Tóth The F.N. David award winner for 2011 is Marie Davidian, North Carolina State http://www.math.washington.edu/~ejpecp University: “For important contributions to the development of methods for analyzing data Electronic Communications in Probability: from longitudinal studies and clinical trials, and for outstanding leadership and dedication to Timo Seppäläinen http://www.math.washington.edu/~ejpecp the statistical profession.” As previously announced, the 2011 Fisher Lecturer was C.Fu Jeff W /ECP/index.php of Georgia Institute of Technology. His lecture was titled “Post-Fisherian Experimentation: Current Index to Statistics: George Styan from Physical to Virtual.” http://www.statindex.org Journal of Computational and Graphical Statistics: We’ll bring you photos and an interview with the Presidents’ Award winner by Bhramar Richard Levine Mukherjee, COPSS secretary, in the next issue. The call for nominations for next year’s http://www.amstat.org/publications/jcgs awards is at http://www.niss.org/copss Statistics Surveys: Lutz Dümbgen http://imstat.org/ss Probability Surveys: Geoffrey Grimmett IMS Fellows at JSM http://imstat.org/ps IMS-Supported Journals Annales de l'Institut Henri Poincaré (B): Thierry Bodineau & Lorenzo Zambotti http://imstat.org/aihp Bayesian Analysis: Herbie Lee http://ba.stat.cmu.edu Bernoulli: Richard Davis http://isi.cbs.nl/bernoulli Brazilian Journal of Probability and Statistics: Silvia Ferrari http://imstat.org/bjps IMS-Affiliated Journals ALEA: Latin American Journal of Probability and Statistics: Claudio Landim http://alea.impa.br/english Probability and Mathematical Statistics: K. Bogdan, M. Musiela, J. Rosiński, W. Szczotka, & W.A. Woyczyński http://www.math.uni.wroc.pl/~pms Other IMS contacts Bulletin website: http://bulletin.imstat.org Main IMS website: http://imstat.org Managing Editor: Michael Phelan [email protected] Ten of this year’s 18 new IMS Fellows were able to travel to Miami for the presentation ceremony. Standing, l–r: Wesley Johnson, Weng Kee Wong, IMS President Peter Hall, Wolfgang Polonik, William Rosenberger, Hongtu Zhu; seated, l–r: Production Editor: Patrick Kelly Maria Eulália Vares, Hal Stern, Peter Bartlett, Jiashun Jin and Chunming Zhang. More photos from JSM are on page 5. [email protected] 4 . IMS Bulletin Volume 40 . Issue 6
Project Euclid news
Project Euclid Receives 2011 PAM Division Award Latest issues of journals on Project Euclid Project Euclid—the premier platform and information community for mathematics and statistics resources from independent pub- IMS journals: lishers—received the 2011 Division Award Annals of Applied Statistics 5(2a) & 5(2b), from the Physics-Astronomy-Mathematics June 2011: (PAM) Division of the Special Libraries http://projecteuclid.org/aoas Association. Given annually, this award (AOAS Vol 5 issue 2b features a special recognizes significant contributions to section on Statistics and Neuroscience) the literature of physics, mathematics, or astronomy, and honors work that demonstrably Annals of Statistics 39(3), August 2011: improves the exchange of information within these three disciplines. The award also takes http://projecteuclid.org/aos into consideration projects that benefit libraries. Annals of Applied Probability 21(3), June “It is an honor for Project Euclid to receive the Special Library Association’s PAM 2011: Award,” said Mira Waller, Project Euclid Manager. “PAM is a community of peers and http://projecteuclid.org/aoap a strong voice for librarians and information experts around the world. In receiving this Annals of Probability 39(4), July 2011: award I feel that Project Euclid is fulfilling its core mission of disseminating scholarly http://projecteuclid.org/aop information in the fields of mathematics and statistics.” Statistical Science 26(2), May 2011: “The PAM Division Award is a real honor,” said David Ruddy, Cornell’s Project Euclid http://projecteuclid.org/ss lead and Director of Scholarly Communications Services at Cornell University Library. “It is also validation that through strong collaborations the academy can effectively address IMS-supported journals: challenges facing scholarly communications.” Annales de l’Institut Henri Poincaré 47(3), Cornell University Library launched Project Euclid in 2000. In 2008, Cornell and August 2011: Duke University Press established a collaborative partnership agreement to jointly manage http://projecteuclid.org/aihp and expand the project. Its mission is to advance scholarly communication in the field Bernoulli 17(3), August 2011: of theoretical and applied mathematics and statistics. It is designed to address the unique http://projecteuclid.org/bj needs of low-cost independent and society journals. Brazilian Journal of Probability and Project Euclid is jointly managed by Cornell University Library and Duke University Statistics 25(3), November 2011 Press. IMS members should already be aware that you can access—free—all issues of [25th anniversary issue—see article all IMS journals via Project Euclid: instructions on setting up access are at http://imstat. on page 16]: org/publications/eaccess.htm. For information about Project Euclid, please contact http://projecteuclid.org/bjps [email protected], or visit projecteuclid.org.
Mentoring for PhD students in developing countries
A new international web-based mentor programme for PhD students in statistics and probability at universities in developing countries has been launched, sponsored by the International Statistical Institute and the Bernoulli Society. PhD students enroll in the programme via the internet, and are put in contact with volunteer senior statisticians and probabilists. The programme is based in Oslo, Norway. See http://statmentoring.nr.no to find out more on how the programme works. We are looking for many volunteer mentors! Senior statisticians and probabilists please enroll at http://statmentoring.nr.no/ statmentoring/index.php/Mentor_form. Membership is free. PhD students in developing countries can join at http://statmentoring.nr.no/ statmentoring/index.php/Student_forms. Please visit the website for more information, and to volunteer or enrol. IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSMCarver IMS@JSM medallist IMS@JSM Ross Leadbetter IMS@JSM (center) IMS@JSM with IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSMIain Johnstone IMS@JSM and IMS@JSM Peter Hall IMS@JSM IMS@JSM IMS@JSMNeyman lecturer IMS@JSM Michael IMS@JSM Jordan (right) IMS@JSM IMS@JSMWald IMS@JSM lecturer Greg IMS@JSM Lawler (right) IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSMExec Committee IMS@JSM et al. IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSMIMS IMS@JSM journal editors IMS@JSM and AEsIMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSMMike Steele IMS@JSM (r) IMS@JSMMedallion IMS@JSM lecture audience IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSMMedallion lecturerIMS@JSM Chris IMS@JSM Holmes (r) IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM Medallion lecturer Michael Newton (l) Medallion lecturer Jianqing Fan (r) IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSMMedallion lecturerIMS@JSM Sylvia IMS@JSM Richardson IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSMIt IMS@JSM wasn’t all hard IMS@JSM work—there IMS@JSM was the IMS@JSM Dance Party, IMS@JSM too. Let’s IMS@JSM hear it: Y–M–C–A! IMS@JSM IMS@JSM IMS@JSM IMS@JSMMedallion lecturerIMS@JSM Qi-Man IMS@JSM Shao (r) IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@ JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM IMS@JSM 6 . IMS Bulletin Volume 40 . Issue 6
Anirban’s Angle: Berkeley & Indian Statistics
in Sankhyá. It is tempting to conjecture that Neyman, that Sankhyá attracted these revolutionary Elizabeth articles because of the Berkeley-ISI personal Scott, bond. Much later, Peter Bickel and J.K. David Ghosh wrote a well-known paper on Blackwell, Bartlett corrections (1990) and the 1982 Peter Bickel, Anirban Blackwell-Ramamoorthi note settled the and Terry Speed DasGupta writes: conjecture that Bayes sufficiency is in gen- have gone to the Recently, I read the 1966 eral weaker than ordinary sufficiency. P.K. ISI; Jeff Wu and Neyman Festschrift volume, edited Sen and J.K. Ghosh contributed an article Jianqing Fan, by Evelyn Fix and F.N. David. In it, to the Neyman-Kiefer proceedings (1985) both Berkeley I found a nice overview of generalized on the LRT for the finite mixture model PhDs, have too. inverses by C.R. Rao, and a lovely article (Hartigan, 1985). On the other hand, Peter Extraordinarily influential work on machine by P.V. Sukhatme, with a long section on Bickel and Anat Sakov contributed a survey learning, high-dimensional inference, bias correction of ratio estimates by sub- on Richardson extrapolation and bootstrap genomics, and random matrices is now sampling. My mind drifted off to the many (Bickel and Yahav, 1988; Politis et al., 1999) going on at Berkeley, and students and fac- other instances of long and productive and Terry Speed and Yee Hwa Yang wrote ulty at the ISI ought to listen to this work connections between Berkeley and Indian on microarrays for the Basu Memorial Issue face-to-face. These would be timely topics statisticians, that have often surpassed the of Sankhyá (2002), all worthy examples of for the Mahalanobis lectures. boundaries of professional collaboration that Berkeley–India connection. A few personal memories. I was Terry’s and have turned into abiding friendships. I Then, too, there has been a steady flow student on his sufficiency course at the ISI, want to reminisce a little, a kind of history of exchange of students and visitors. B.V. and just this year he and I worked on put- looking forward. Rao was invited to Berkeley after he solved ting together Basu’s most influential work Early instances that I recall are the Ulam’s problem in his thesis. Ashok Maitra in a Springer volume. It was a lovely period PhD dissertation of Ashok Maitra under visited Berkeley several times. Few know of my life. Sandrine Dudoit and I just David Blackwell, Prem Puri with Neyman, that Terry Speed went to ISI and taught for worked on a survey of sufficiency. David Sudhakar Dharmadhikari with Barankin, ISEC, the international wing of ISI. Three Freedman was advising me on models for and D. Basu’s voyage to Berkeley on a of Peter Bickel’s earliest students were Hira fractional data up until six days before his Fulbright in 1953. The magnificent Maitra- Koul, D.P. Gokhale, and R.K. Aiyar. Erich tragic death. I first listened to Peter Bickel Sudderth text on Discrete Games (1996) Lehmann had (nearly) countless Indian in 1980 at a conference at the Delhi ISI. is clearly influenced by David Blackwell’s students—Gouri Bhattacharya, M.L. Puri, The responsible ISI official greeted a large seminal contributions to stochastic games. M. Raghavachari, to name but a few. Rabi contingent of us at the Calcutta central And Basu’s entire professional development Bhattacharya did some of his most influen- train station with a confident toothy smile was directly influenced by his associa- tial work at Berkeley. Most recently, Antar and said, “Board this train.” Some twenty tion with Neyman, and later with David Bandyopadhyay, Smarajit Bose, Probal minutes later, we were all detained for Blackwell and Lucien Le Cam. Chaudhuri, and Manjunath Krishnapur ticketless travelling and given a hefty fine. Some other early examples of the have returned to India after doctoral work On our return, ISI reimbursed us for the Berkeley–India relationship are the articles at Berkeley, and Ani Adhikari and Sourav fines, but it had to be shown as taxi fare. by Rao, Bahadur, and Basu in the Berkeley Chatterjee are current members of the Peter was also taking the train from New symposium proceedings (1965). Although Berkeley faculty. Such a long history! Delhi to Calcutta; he was a little late, and not a joint collaboration, the Rao-Blackwell Visits, seminars, and personal conversa- at the stairs, I only had time to shake his theorem is a household name, and such tions are extremely helpful for exchange of hand. I properly met Peter for the first time cornerstones of inference such as the ideas, confirmation, falsification, and crys- in 1991 at a conference in Ottawa: I gave Lehmann-Scheffé theorem were published tallization of what is only a thought. I know a very simple talk on extremum efficiencies September . 2011 IMS Bulletin . 7
in some nonparametric problems and he At Purdue, I learned of the close profes- loss for me. He was a friend.” made some comments to me after the talk. sional and personal relationship between That’s what I mean: there is a long and Peter invited me to come to Berkeley that Shanti Gupta and Peter Bickel and Lucien treasured friendship between Indian statis- Fall, and I spent a month at the MSRI. Le Cam. I can vouch that Shanti earnestly tics and Berkeley, and I so much wish to see I was interested in some problems on counted on counsel from Peter and Lucien it prosper and last. ■ convolutions at that time and I recall Peter Le Cam. In the January of 2002, Shanti coming to my office at Evans Hall to help passed away most unexpectedly after a brief What about other special working relation- me. The little work on t-intervals, Basu and illness. I informed Peter the same evening. I ships? Do you have a story to share? You DasGupta (1995), also benefitted from that recall Peter writing back “This is a great loss can comment on this (and other articles) at MSRI visit. for Purdue. I am very sorry. It is a personal http://bulletin.imstat.org. Go on!
I saw This the Laha awards on was my first IMS Laha Awards 2011 the IMS website and my advisers meeting. I learned a lot from This was encouraged me to apply. It's a great pleasure others’ research presentations, and the first time I attended to receive this award! One suggestion: I think met lots of interesting people. Miami is the Presidential Address, and I won’t it may get more attention from us students if a a great place, I liked the beach! I would forget the moment I was called as a Laha presentation session (or two) is organized for definitely encourage people to apply for Award recipient. The highlight of the evening was Laha award recipients in the future. next year’s Laha awards. meeting many statistical fellows, including IMS Gongjun Xu, Columbia Dan Shen, UNC Chapel president Peter Hall. I really enjoyed JSM. University Hill Hui Nie, University of Pennsylvania 8 . IMS Bulletin Volume 40 . Issue 6
Obituary: Anatolii Skorokhod 1930–2011
The world’s mathematical community Doctoral dissertation, SDE and limit theo- lost one of its most prominent members rems for stochastic processes, and was made when Professor Anatolii Vladimirovich Professor of the Kyiv State University the Skorokhod died on January 3, 2011. following year. A year later he moved to the Anatolii Skorokhod was born on September Institute of Mathematics of Academy of 10, 1930 in Nikopol, Ukraine. In 1948, Sciences of Ukraine as Head of Department Anatolii V. Skorokhod in 2003 (photo: Alex Novikov) he graduated from secondary school of the theory of stochastic processes. He (with a gold medal) and entered the Kyiv became a Corresponding member of the examples of continuous functionals. State University, where he specialized in Ukrainian Academy of Sciences in 1967 The most well-known topology is the probability theory at the Department of (and an Academician in 1985), and was J1-topology (we call it the Skorokhod topol- Mathematical Analysis, under the consider- awarded the Ukrainian State Prize in ogy), which proved to be very useful in the able influence of Professors B.V. Gnedenko Science and Technology in 1982 and 2003. case of typical path spaces. and I.I. Gikhman. In one of his first papers In 1985 Skorokhod moved to Michigan A.V. Skorokhod obtained many sig- (1954) Skorokhod clarified the structure of State University, USA, and in 2000 he was nificant results in the theory of stochastic densities of stable distributions. elected to the American Academy of Arts differential equations, namely: the theorem In the early 1950s, the atmosphere at and Sciences. of comparison for linear diffusion, which Kyiv University was far from friendly, which A.V. Skorokhod’s contributions to allowed him to get the first results on the resulted finally in B.V. Gnedenko’s depar- probability theory, mathematical statistics, existence of strong solutions for the case ture. Foreseeing this, Gnedenko wrote to and the theory of stochastic processes, were of non-Lipschitz drift coefficients; the A.N. Kolmogorov: “Dear Andrei, one more fundamental, in particular in the areas: theory of diffusion processes with boundary matter. Here we have a graduating student, limit theorems for stochastic processes; conditions of different types; asymptotic very able in my view, Skorokhod Anatolii stochastic differential equations; probability behavior of solutions; fundamental results Vladimirovich, 22 years of age. Things are distributions in infinite dimensional spaces; on ergodicity and stability of solutions. We going in such a way that it would be better for stochastic integration of anticipating func- also mention his results on the problem of him to leave and attend postgraduate courses tions; and asymptotic behavior of dynamic differentiability of probability measures; the elsewhere. If there is a possibility, by the systems under random perturbations. structure of the Radon-Nikodým derivatives ministry, to assign him to Moscow University In the 1950s A.V. Skorokhod began for quasi-invariant measures on a Hilbert for postgraduate study, then you will do a good a large cycle of works on limit theorems space. deed both for the student and for science...” for stochastic processes. He proposed an Along with a great number of scientific (From a letter of B.V. Gnedenko to A.N. original method of studying limit theorems, papers (about 300), A.V. Skorokhod wrote Kolmogorov, October 30, 1952.) based on two profound ideas: the method many books, textbooks, and monographs Thus, by the efforts of Kolmogorov, of a single probability space and introduc- (more than 25), which are well-known in Moscow State University acquired three tion of several topologies on the space of the probability world and are constantly post-graduate students: A. V. Skorokhod functions without discontinuities of the used both by students and experts in prob- (supervisor E.B. Dynkin), V.S. Korolyuk second kind. His main idea in the area of ability theory and the theory of stochastic and V.S. Mikhalevich. All of them would limit theorems was to introduce a topology processes. His last book, with Habib Salehi later become Members of the Academy of which would make this space a Polish space and Frank Hoppensteadt, was Random per- Sciences of Ukraine. (complete separable metric space). Along turbation methods with applications in science A.V. Skorokhod wrote his Candidate this line A.V. Skorokhod introduced four and engineering (Springer, 2002, Ser. Appl. dissertation, Limit theorems for random topologies; established necessary and suf- Math. Sci., vol. 150). processes, at Moscow State University, ficient conditions for convergence in those Alex Novikov (University of Technology, in 1956, and returned to the Kyiv State topologies; described compact sets for each Sydney, Australia) and Albert Shiryaev University. In 1962 he completed his topology; and listed the most important (Steklov Mathematical Institute, Moscow) September . 2011 IMS Bulletin . 9
Obituary: Patrick Billingsley 1925–2011
Patrick Billingsley, scholar, writer, and actor, David Huntsberger, 1986), Ergodic Theory died Friday, April 22 at the age of 85. He is and Information (1965), Convergence of survived by his children Franny, Patty, Julie, Probability Measures (1968), and Probability Marty and Paul, and by his companion, and Measure (1978). The last three all Florence Weisblatt. His late wife of nearly became, at least for a time, the authorita- 50 years, social activist Ruth Billingsley, tive works on their subjects; all are still Patrick Billingsley died in 2000. widely cited, and Probability and Measure, Billingsley was born May 3, 1925 in now in its third edition, is still a standard chaotic dynamical systems in the 1970s and Sioux Falls, South Dakota. He earned graduate-level textbook. Billingsley also 1980s, which culminated in the celebrated a bachelor’s degree in engineering from published a number of expository papers, Ledrappier-Young formula expressing the US Naval Academy in 1948, then including an article on prime numbers and entropy as the sum of the Lyapunov served in the US Navy until 1957. As a Brownian motion that was awarded a Lester exponents multiplied by the directional Navy officer he lived for a year in Japan, R. Ford prize for mathematical writing. In Billingsley dimensions. Billingsley also where he earned a black belt in judo. this article he explained that his approach made important contributions to the field While still a naval officer he entered the to mathematical exposition had come from of weak convergence—beginning with his Princeton University graduate program in his PhD mentor Feller: “For the most part PhD thesis on the invariance principle mathematics; he received a master’s degree I shall only illustrate general results by for dependent random variables—and to in 1952 and a doctorate in 1955, the latter examples and special cases. For this there is probabilistic number theory. He delivered under the supervision of William Feller. In the authority of William Feller, who used the 1973 Wald lectures on the probability 1957–58 he worked as a National Science to tell us, his students, that the best in theory of additive arithmetic functions, and Foundation Fellow in Mathematics at mathematics, as in art, letters, and all else the 1972 Rouse Ball lecture at Cambridge Princeton. Billingsley joined the faculty […] consists of the general embodied in the on prime numbers and Brownian motion. at the University of Chicago in 1957, concrete.” In the mid-1960s Billingsley embarked with appointments in the departments of He went on, perhaps recalling his days on what became a parallel career in acting. Statistics and Mathematics, and remained in the US military, “Although at first I His first roles were in amateur productions, on the faculty until retiring as professor thought that was simply an anti-military but beginning in 1970 he played leading emeritus in 1994. He served as chairman statement, I did eventually understand roles in more than 20 professional produc- (or, as he liked to put it, “acting chair- it as the intellectual-esthetic principle he tions at the Court Theatre and Body Politic man”) of the Statistics department, from intended and have tried ever since to keep it Theatre in Chicago. His roles included 1980 to 1983. He was a Fulbright Fellow at the front of my mind.” the Captain in We Bombed in New Haven and visiting professor at the University of Billingsley is perhaps best known in (1970); Alonzo in The Tempest (1977); Copenhagen, Denmark in 1964–65 and a mathematical analysis as the inventor of Dysart in Equus, (1980); and Petey in The Guggenheim Fellow and visiting professor what is now called Billingsley dimension, Birthday Party (1978 and 1985). A talent at the University of Cambridge, England, an extension of the notion of Hausdorff scout saw him perform in The Lover in in 1971–72. Billingsley was the first editor dimension to positive Borel measures. 1977, and this led to a successful audition of the Annals of Probability, from 1976 to The Billingsley dimension is one of three for a part in the 1978 Kirk Douglas film 1979, and President of the IMS in 1983. fundamental quantities attached to an The Fury. Billingsley never met Douglas, Billingsley’s professional accomplish- ergodic, invariant probability measure in but they appear on screen together during ments were many and varied, but he will be a smooth dynamical system, the others a car chase on Wacker Drive and Van remembered first and foremost as an exposi- being the Kolmogorov-Sinai entropy and Buren Street in Chicago; Billingsley played tor. He wrote five books, Statistical Inference the Lyapunov exponent(s). The relation- a bad guy who ended up dying in a fiery for Markov Processes (1961), The Elements of ships among these three quantities was an crash. He went on to appear in seven more Statistical Inference (in collaboration with important theme in the study of smooth, films and in nine television shows. Among Continued on page 11 10 . IMS Bulletin Volume 40 . Issue 6
Rick’s Ramblings: Writing an NSF proposal with New Broader Impacts
Being an accomplished Jedi Master, I feel a sense of noblesse oblige when it comes to instructing the Padawans in the ways of the Rick Durrett, Jedi Master of proposal- Force. Today’s column is written primarily for a Youngling who writing, writes some instructions for the intends to submit an NSF proposal to the probability program at Younglings, including the NSF’s revised NSF for the November 7, 2011 deadline. Undoubtedly much of broader impacts criterion. this advice applies to statistics proposals and to grants submitted to other agencies, but in that case you should heed the advice people have given me for years: get professional help. Those competitions require a different sort of verbal fertilizer, so you should seek advice from a colleague who has had success with that type of grant. of academic year support plus benefits plus tuition (and overhead) The first and most important rule for writing a proposal is to adds up quickly. It is easy to let your budget fantasies get the best get your materials ready well before the deadline. This is neces- of you, so keep them in check. Gayle, who processed my grants sary because you don’t actually submit your grant. It is sent in by for over 10 years at Cornell, had a cartoon on her door with the your Office of Sponsored Projects. At most universities, after you punch-line, “What the heck, I’ll add another zero.” have uploaded your proposal to Fastlane, your department’s grants NSF grants are evaluated based on two criteria: “intellectual person looks it over and when she is satisfied it is complete, you merit” and “broader impact” introduced in 1997. These MUST be release to OSP to view, edit, and submit. After OSP has verified explicitly addressed in the proposal summary, and should be done that you are following the university’s rules about tuition, indirect in two paragraphs that start with the two phrases in bold face. To costs, and cost sharing then it is off to NSF. quote http://www.nsf.gov/pubs/gpg/broaderimpacts.pdf: “Experience A generic NSF proposal has five main parts. A one-page Project shows that while most proposers have little difficulty responding to the Summary, a 15-page Project Description, References (with no page criterion relating to intellectual merit, many proposers have difficulty limit), a two-page NSF style CV, and a budget, with up to three understanding how to frame the broader impacts of the activities they pages of explanation. The CV is the easiest so I’ll start with that. propose to undertake.” This document has a large number of useful The first thing you need to do before submitting a proposal is to go suggestions for things to include, such as: service as referee or online and get the Grant Proposal Guide, which is now a hefty 71 associate editor, IMS or AMS committees, conference organization, pages. It is not a great read but fortunately it has a detailed table of interdisciplinary research activities, course development, under- contents. graduate research supervision, and high school outreach. In Section C.1.f of Chapter 2 you will find precise instructions The broader impacts criterion has recently been revised. for your CV. How to describe your education, previous positions, OMG! (as the Younglings might say). Letters in the July 8 Science up to five publications related to the proposal, up to five significant (volume 333, 157–158) and in the July 14 Nature (volume 475, publications, up to five synergistic activities, all your collaborators page 141) rail against the changes, saying “they move too far in the in the last 48 months, all your PhD students ever, and recent direction of accountability at the cost of scientific creativity and postdocs mentored. You shouldn’t sweat the limit on relevant publi- autonomy,” and calling the goals “at best arbitrary and at worst cations. You can refer to as many of your papers as you want in the an exercise in political triangulation.” In brief, the new National proposal narrative and list them in the references. Science Board plan, posted online June 14 (see nsf.gov/nsb/ The second easiest piece is the budget. For the youngest investi- publications/2011/06_mrtf.jsp), will require researchers to identify gators this is simple: two months’ summer salary, and some money their broader impacts by selecting from a list of nine national for travel or perhaps a new laptop. If your department demands priorities: you ask for some money for supplies, then go along, but forget • Increased economic competitiveness about page charges unless you publish in biology journals where • Increased national security they don’t take no for an answer. If you are at the point of having • Increased partnerships between academia and industry graduate students then think about summer support, but the cost • Enhanced infrastructure for research and education September . 2011 IMS Bulletin . 11
• Development of a globally-competitive or samples that require management and/or going to work on but, then again, your STEM workforce sharing. PIs should note that the statement proposal will be evaluated by panelists who • Increased participation of women, will be subject to peer review.” are generally knowledgeable about your persons with disabilities and underrepre- Climbing down from my soapbox, it research area but are not experts on your sented minorities is now time to address the nitty-gritty of particular research topic: a probability panel • Improved pre-K to 12 STEM education writing the proposal narrative. If you will have between 10 and 12 people who and teacher development haven’t started yet, then you are three have to review 90 proposals. • Increased public scientific literacy months late. About six months before the The final thing to say is: don’t get The first two were red flags for the bulls in deadline you should start jotting down ideas discouraged. Serving on the probability Science who say the list “excludes protecting of things to work on. You can’t write a plan panel is very depressing. Only about 10% the environment and addressing other social for three years of research in two weeks, no of the proposals are truly “not worthy of problems” and this may “undermine the matter how many hours a day you sit there support,” but only about one-third can be attractiveness of STEM disciplines to more and scratch your head. You need time to sit funded. This situation leads to some agoniz- idealistic students, who are interested in and think, scribble exploratory calculations ing and occasionally random decisions. So, meeting human needs rather than fostering on your tablet, and read articles about new if you are just starting out, keep in mind economic competitiveness.” A little melo- things you might work on. that it will take several tries to have success. dramatic, I think. I am all for protecting People who have already had NSF Read the reviews carefully, discuss them the environment—e.g., not drilling for oil grants will begin their proposals with up with your colleagues and keep in mind their off the outer banks of North Carolina—but to five pages of results from prior support, assessment represents personal opinions that shoving in Republican faces the fact that which highlight previous achievements and my change dramatically from year to year. ■ NSF science can stand in the way of corpo- make the case that they can actually do rate America screwing the environment for what they say. Even with up to five pages profit is not going to help the budget. used up with this, there is room to discuss Patrick Billingsley, 1925–2011 Oh well, even if you hate it, the simple 4–6 problems, which need to be explained Continued from page 9 fact of the matter is that unless you address in sufficient detail and related to the other roles, he played a biology teacher the broader impacts of your research, as literature. in My Bodyguard (1980), the professor in well as jump through the new hoops of the The problems you pose must be inter- Somewhere in Time (1980) and the bailiff in new Postdoc Mentoring Plan and the Data esting but accessible. It is one thing to say The Untouchables (1987). Management Plan, your proposal will be that you will use the winding of Brownian Billingsley was, in the words of his dead on arrival. Most first timers will not motion in the infinitely punctured complex daughter Marty, a “true Renaissance man”, be asking for postdoc support, so the PMP plane to prove the Riemann Hypothesis, a man of diverse interests who excelled in is moot, but everyone is forced to have a but unless you are Stas Smirnov, you will many things. He painted; he wrote; he read DMP even if it consists of the simple state- need to work hard to convince the proposal Beowulf in the original Old English; he ment “Data, I don’t need no stinking data.” reviewers that you have a chance of doing worked out daily in the university gym for In less flippant terms, the NSF FAQ says “It this. On the other hand the problems you nearly 40 years. He had a wry comment for is acceptable to state in the DMP that the propose should not be too close your previ- almost every occasion. When he vacated project is not anticipated to generate data ous work. In my case if I were to write that his office in Eckhart Hall some years back, I was going to use the block construction he invited me to help myself to whatever “The Force is strong in this NSF grant proposal” to prove coexistence for some interacting books were there. I was amused—but particle systems that arose from ecological not surprised—to find, between a worn competition then I would expect reviewers copy of Burns’ Poems and Songs and a to complain. Complete Works of William Shakespeare, Younglings, you should not be ashamed a copy of Arnold Schwarzenegger’s 1978 to ask for help from more experienced autobiography, Arnold: The Education of a colleagues. Have them read over what you Bodybuilder. F have written and make comments. They Steve Lalley, University of Chicago
Photo: likr/reway2007 may not know the details of what you are 12 . IMS Bulletin Volume 40 . Issue 6
“Ranking Our Excellence,” or “Assessing Our Quality,” or Whatever…
One of the final responsibilities of each IMS President is to deliver a Presidential Address at the IMS annual meeting. Peter Hall gave his talk on August 1 at JSM in Miami. The text is reproduced in full here.
e live in an era where almost everything apparently can be quantified, and most things are. Massive quantities of data are generated every day on subjects ranging Wfrom our supermarket purchases to changes in the climate. This creates unprecedented opportunities for statisticians, but also many challenges. Not least are the challenges of quantifying our own professional lives. The managers of the institutions, organisations and companies where we work measure our performance, and the quality of our work, using rankings, bibliometric analyses and similar approaches. Tonight I want to say a few words about these matters. They are analysis. An extensive manual describes methodology and metrics. fundamentally statistical in many ways, yet no-one, least of all us Admirably, last November the ASA organised a NISS workshop on or our managers, has many of the tools necessary to undertake the assessing the quality of graduate programs. However, the ranking of analysis properly and convey the results. In many cases we do not graduate programs does not meet the criterion of transparency. As a even seem to have the knowledge needed to develop the tools. result, the ranking sometimes has been misinterpreted by university Rankings, for example of the institutions or departments where managers. More generally, statistical issues relating to modelling we work, are among the most common of the techniques that are rankings, describing their authority and simplifying their interpre- used to analyse us, and perhaps to divide and conquer us. Goldstein tation, deserve greater attention from us than we have given them. and Spiegelhalter, in a paper in the Journal of the Royal Statistical Rankings, like bar graphs and pie charts, are only a way Society (Series A) in 1996, argued that no ranking should be unac- of presenting data. The data themselves should also be a focus companied by a measure of its authority. This recommendation is of our attention. The IMS as a society, and also through its seldom honoured. Rankings should also be reasonably transparent, individual members, has played an important role in the study of so that the implicit statements that they make about us are clear to performance-related data. In particular, the IMS was one of three non-experts (in some cases that includes us!). professional bodies (the others were the International Mathematical International rankings of universities are a case in point. The Union and the International Council of Industrial and Applied data on which they rely are seldom accessible, and the methodol- Mathematics) which produced the report Citation Statistics, ogy they use is typically secret. The methodology usually can be addressing that most pernicious of academic topics, bibliometric accessed only approximately, for example via a process of reverse data and their analysis. The report’s authors, Robert Adler, John engineering, noting the changes in rankings after Nobel Prizes, etc, Ewing and Peter Taylor, expressed concern about the use of citation have been awarded. For all these reasons the rankings are far from data for assessing research performance. Their report, published in transparent. And they are seldom accompanied by a measure of Statistical Science in 2009, noted that: their reliability. Yet they influence significantly the comments, and subsequently the actions, of university managers and, sometimes, There is a belief that citation statistics are inherently more the decisions of governments. accurate because they substitute simple numbers for complex judg- The US National Research Council ranking of university statis- ments, and hence overcome the possible subjectivity of peer review. tics departments, compiled using data on graduate programs and released almost a year ago, is to be applauded for incorporating sta- Adler, Ewing and Taylor pointed to the fallacy of such beliefs, and tistical measures of authority, based on resampling ideas, into their drew conclusions that, although concerning to some scientists (and September . 2011 IMS Bulletin . 13
to university and other institutional managers), resonate with many but while the correlation between peer assessment and biblio- of us: metric indicators is statistically significant, it is not perfect. • The accuracy of citation metrics (e.g., raw citation counts, (ii) “Bibliometric indicators may be considered as approximation impact factors, h-factors and so forth) is illusory. Moreover, the measures of the inherent quality of papers, which, however, misuse of citation statistics is widespread and egregious. In spite remains fully assessable only with aid of human unbiased judge- of repeated attempts to warn against such misuse (e.g., the mis- ment, meditation, and elaboration. We advocate the integration use of the impact factor), governments, institutions, and even of peer review with bibliometric indicators, in particular those scientists themselves continue to draw unwarranted or even false directly related to the impact of individual articles, during the conclusions from the misapplication of citation statistics. next national assessment exercises.” • Sole reliance on citation-based metrics replaces one kind of Australian authorities too have been endeavouring to turn biblio- judgment with another: Instead of subjective peer review one metric data to good use, so as to distribute block grant funding to has the subjective interpretation of a citation’s meaning. universities. (This is the type of funding that, in the US, is passed • While statistics are valuable for understanding the world in on largely in the form of grant overheads. In this sense, grant which we live, they provide only a partial understanding. Those income in the US is used as a proxy for research performance.) who promote the use of citation statistics as a replacement for Interestingly, in 2011 a trial run of Australia’s research assessment a fuller understanding of research implicitly hold such a belief. exercise led the federal government to conclude that a rather We not only need to use statistics correctly—we need to use controversial ranked list of journals, to which the government had them wisely as well. seemed to be committed, was not providing the overall benefits Bibliometric analyses, and other statistical measures, are sometimes that had been anticipated. Announcing the virtual abandonment used in connection with appointment and promotion cases, of journal rankings three months ago, the Australian Minister for occasionally with dramatic consequences (as we’ll see below, when Innovation, Industry, Science and Research noted that the ranking discussing the Australian experience). Indeed, potential applica- was being seriously misused by university managers: tions to appointment and promotions decisions are major reasons for heightened interest in the interpretation and application of There is clear and consistent evidence that the rankings were being bibliometric analyses. However, many other empirical approaches deployed inappropriately within some quarters of the [university] have been employed. They range from the Texas A&M University sector, in ways that could produce harmful outcomes, and [were] system’s development of methodology, which evaluates how much based on a poor understanding of the actual role of the rankings. university professors “are worth,” based on their salaries, how much One common example was the setting of targets [in connection research money they bring in, and how much money they generate with appointments or promotions] for publication in [highly from teaching; to the much more widespread use of student evalu- ranked] journals by institutional research managers. ations of classroom teaching. All have weaknesses, and some have strengths as well. However, there is concern that in a range of research fields in Several nations have attempted, or are currently attempting, to Australia, rankings based on citation analyses, rather than peer unlock the secrets of bibliometric data so that they can use them review, will fill any gaps left by reducing reliance on journal rank- to their advantage. For example, the first Italian national research ings. Moreover, journal rankings will still apparently be used to evaluation, which commenced in 2003, sought to answer the fol- some extent in the Australian research assessment process. lowing questions, among others: The Australian government’s enthusiasm for a research assess- (i) Are peer review judgements and bibliometric indicators inde- ment exercise that emphasises bibliometric analysis, and, in many pendent variables, and if not, what is the strength of association fields of research, seems to give peer review a significantly lesser between them? role, would appear to be at odds with the Italian experience. It (ii) Is the association between peer judgement and article citation is also in conflict with the UK’s proposed new methodology for rating significantly stronger than the association between peer assessing research importance, which, as we shall now relate, judgement and journal citation rating? originally seemed to favour a largely bibliometric process but then The Italians concluded that: retreated. Just as importantly, the Australian government’s decision (i) Bibliometrics are not independent of peer review assessment; to largely abandon journal rankings reflects an issue that has to 14 . IMS Bulletin Volume 40 . Issue 6
be borne in mind whenever funding is distributed as the outcome The UK apparently has withdrawn from this position. The Research of a process of performance evaluation: As is no doubt hoped, Excellence Framework (REF), which will replace the RAE and be the people receiving that funding respond to their evaluation by run for the first time in 2013, will instead assess research in at least changing their behaviour so as to achieve more funding next time, three ways and incorporate peer review: but they may change in ways that are overtly counterproductive to • Outputs (total weight 65%): The primary focus of the REF achieving the goals that the process was originally designed for. exercise will be to identify excellent research, apparently using The UK once argued rather forthrightly that its Research largely expert peer review but, in subjects where robust data are Assessment Exercise (RAE) should not explicitly take account of available, “peer review may be informed by additional citation citation data and the like. For example, the instructions to 2008 information.” RAE panels included the following injunction: • Impact (20%): The REF will endeavour to identify cases where “researchers build on excellent research to deliver demonstrable In assessing excellence, the sub-panel will look for originality, benefits to society, public policy, culture, quality of life and the innovation, significance, depth, rigour, influence on the discipline economy.” and wider fields and, where appropriate, relevance to users. In • Environment (15%): The REF will assess, and take into assessing publications the sub-panel will use the criteria in normal account, the quality of the research environment. use for acceptance by internationally recognised journals. The In the UK, particularly in the mathematical sciences including sub-panel will not use a rigid or formulaic method of assessing statistics, the second of these criteria has become the most con- research quality. It will not use a formal ranked list of outlets, nor troversial, not least because of its potential to focus research on impact factors, nor will it use citation indices in a formulaic way. short-term goals. It is admittedly less of a problem for statisticians than for, say, pure mathematicians. However, the replacement for the RAE was originally intended to The US is not as vulnerable to these difficulties as many other use metrics in place of peer review: nations—such as the UK, European countries including France, Italy and The Netherlands (but not Germany), and Australia— It is the Government’s intention that the current method for that have relatively unified national higher education systems. determining the quality of university research—the UK Research Admittedly, substantial research funding is provided federally in Assessment Exercise (RAE)—should be replaced after the next the US, but few US states would allow their university systems to cycle is completed in 2008. Metrics, rather than peer-review, will be scrutinised federally in the manner that is accepted practice in be the focus of the new system and it is expected that bibliometrics other countries. In a state or provincial system a state can experi- (using counts of journal articles and their citations) will be a ment with different ways of measuring and rewarding research central quality index in this system. performance, and the others can sit on the sidelines and watch the [Evidence Report, Research Policy Committee of Universities, p.3] experiment, adopting the methodology only if it is effective. As statisticians we should become more involved in these mat- The Times Higher Education Supplement for 9 November 2007 ters than we are. We are often the subject of the analyses discussed headlined a story on its front page with the words, “New RAE above, and almost alone we have the skills to respond to them, for based on citations,” and commented thus: example by developing new methodologies or by pointing out that existing approaches are challenged. To illustrate the fact that issues … After next year’s RAE, funding chiefs will measure the number that are obvious to statisticians are often ignored in bibliometric of citations for each published paper in large science subjects as analysis, I mention that many proponents of impact factors, and part of the new system to determine the allocation of more than other aspects of citation analysis, have little concept of the prob- £1 billion a year in research funding. A report published by lems caused by averaging very heavy tailed data. (Citation data are Universities UK enforces such a “citations per paper” system as typically of this type.) We should definitely take a greater interest in the only sensible option among a number of so-called bibliometric this area. ■ quality measurements. It concludes that measuring citations can accurately indicate research quality. Acknowledgement: I am grateful to Rudy Beran, Valerie Isham and Wolfgang Polonik for helpful comments. September . 2011 IMS Bulletin . 15
SPA 2011, Oaxaca, Mexico
Mireille Chaleyat-Maurel, Université Paris Descartes, Paris, France, networks. The Lévy lecture by Jean-François Le Gall described the reports from June’s Conference on Stochastic Processes and their solution of a long-standing problem concerning scaling limits of Applications: What ingredients make a successful conference? An random triangulations of the sphere. The Itô prize was awarded to attractive scientific program, an efficient local organizing committee Haya Kaspi and Nathalie Eisenbaum for their paper on permanen- and a beautiful location. All these were present in Oaxaca for the tal processes published in the journal Stochastic Processes and their SPA meeting! Applications in 2010. In addition, there was a large variety of special The 35th Conference on Stochastic Processes and their sessions, consisting of three talks each, contributed session talks, Applications was organized under the auspices of the Bernoulli and posters. Society and co-sponsored by the Institute of Mathematical The social program was sensational. On Monday, a reception Statistics. It took place in the city of Oaxaca, Mexico from the 19th was offered at the Museum of Contemporary Art of Oaxaca. to the 24th of June 2011. The SPA conference is the major annual Besides exhibiting Modern Art from the whole world, the Museum meeting for researchers working in the field of Stochastic Processes periodically shows works by Oaxacan master painters Rufino and their Applications. The Oaxaca conference gathered 275 par- Tamayo, Francisco Toledo, Rodolfo Morales, Rodolfo Nieto and ticipants from 32 different countries. Francisco Gutiérrez. The conference was hosted by the Oaxacan branch of the Wednesday afternoon featured an excursion to the very interest- Instituto de Matemáticas of the Universidad Nacional Autónoma ing location of Monte Albán, a large pre-Columbian archaeological de México and was co-organized by several Mexican research site near Oaxaca. Monte Alban was the ancient capital of the centres in probability. Zapotecs and one of the first cities in Mesoamerica. It was founded The organization of the Conference was perfect. As soon as approximately 500 years BC and flourished until 750 AD. participants arrived at Oaxaca Airport (sometimes very late in The conference dinner was in the historic “Ex-Convento de the evening), they were met by members of the local organizing Santa Catalina” with a show of the Guelaguetza, (or Mondays Committee who took care of them, and helped arrange transport of the Hill) with traditional dancing in costume in groups. Each to their hotel. On Sunday, the registration at Hotel Mision Los costume and dance had local cultural significance. Angeles was followed by a small reception with very good tequila (Tequila Sauza was one of the sponsors of the Conference!). Conference material was provided to every participant in a nice Mexican bag (featuring a design of the celebrated Oaxacan painter Francisco Toledo) and contained all desir- able information about the program and the city of Oaxaca. Let me add that all the contents (even the pen!) were recyclable. Rivero The 20 invited plenary lectures were V presented by leading experts in their fields and took place in the Teatro Juarez, near to Photo: íctor Traditional dancers entertained the participants at the conference dinner the Hotel Mision. These lectures covered a wide range of active research areas: random trees and graphs, perco- In addition, the city of Oaxaca, was very safe and offered a great lation, statistical mechanics and the Ising model, random matrices, variety of interesting places where the participants could rest after stochastic networks, control theory, stochastic differential equations the conference. and more (the full program is on the conference website). Warm thanks to the local organizing committee, who did a The program featured two IMS Medallion lectures, by Itai wonderful job under the coordination of María Emilia Caballero Benjamini on recurrence problems for random walks, and Alice and Víctor Rivero. Guionnet on random matrices. Ruth Williams delivered the Doob The conference website (with program details) is located at lecture and reported on recent results on the analysis of stochastic http://www.matem.unam.mx/~SPA2011/ 16 . IMS Bulletin Volume 40 . Issue 6
Brazilian Journal celebrates 25th anniversary
The Brazilian Journal of Probability and Cordeiro. Previous theory and methods edi- years to come. Statistics (BJPS) is an official publication tors are Gauss M. Cordeiro, Bent Jørgensen “This anniversary issue contains papers of the Brazilian Statistical Association and and Chang C. Y. Dorea, and previous by outstanding researchers. They cover is supported by the IMS. Currently three applications editors are Julio M. Singer and important topics in probability and statis- issues a year, with four planned, the journal Jorge Achcar. (Anthony Davison was theory tics. The papers below [see box]make up the publishes papers in applied probability, and methods editor for a brief period, prior current BJPS issue. applied statistics, computational statistics, to assuming the editorship of Biometrika.) “We would like to take the opportunity mathematical statistics, probability theory We took over the journal in 2007. In the to thank the authors of the papers that and stochastic processes. Subscription is following year, an agreement between the make up this anniversary issue. We also $100 for IMS members. See http://www. Brazilian Statistical Association and the thank all previous editors for their hard imstat.org/bjps/subscriptions.html Institute of Mathematical Statistics (IMS) work, the Brazilian Statistical Association, BJPS is marking its quarter-century was established which resulted in the BJPS the partial funding received from with a special anniversary issue (Vol. 25, being an IMS-supported journal (since CNPq and CAPES, and the Institute of No. 3, 237–238). The Preface of this issue, 2008). This change resulted in a faster and Mathematical Statistics. The publication reproduced with permission, reads: more convenient refereeing process through of the current special issue was partially “The current issue celebrates the 25th an electronic management system of sub- funded by the INCTMat grant, and for anniversary of the Brazilian Journal of missions and also in wider indexation. As a that we thank Professor Jacob Palis. Finally, Probability and Statistics, an academic consequence, in 2010, the number of issues a special recognition goes to Professor journal published by the Brazilian Statistical in each volume was increased from two to Wilton Bussab (in memoriam), who was Association since 1987. The journal three. These are published in March, July the president of the Brazilian Statistical publishes high-quality research papers in and November. Our goal is to publish four Association when the agreement with the applied probability, applied statistics, com- issues per year, which is expected to happen IMS was signed.” putational statistics, mathematical statistics, in the near future. In April 2011, BJPS Silvia L. P. Ferrari (Chief editor); probability theory and stochastic processes. was included in the Scopus bibliographic Francisco Cribari-Neto (Applications Previous chief editors are Pedro A. database. We firmly believe that the BJPS editor); and Nancy L. Garcia (Theory and Morettin, Heleno Bolfarine and Gauss M. will continue its pursuit of excellence in the Methods editor)
Contents of BJPS special anniversary issue (Volume 25, Issue 3) Prediction-based estimating functions: Review and new developments, Additive models for quantile regression: Model selection and confidence by Michael Sørensen bandaids, by Roger Koenker Local linear suppression for wireless sensor network data, by Kristian Dispersion models for geometric sums, by Bent Jørgensen and Lum and Alan E. Gelfand Célestin C. Kokonendji Modelling particles moving in a potential field with pairwise Stationary infinitely divisible processes, by Ole Eiler Barndorff- interactions and an application, by David Brillinger, H. K. Preisler Nielsen and M. J. Wisdom
Limit theorems for empirical Fréchet means of independent and non- Hierarchical wavelet modelling of environmental sensor data, by identically distributed manifold-valued random variables, by Wilfrid Anthony Davison and Yann Ruffieux Stephen Kendall and Huiling Le On improved estimation for importance sampling, by David Firth On default priors and approximate location models, by Donald Fraser and Nancy Reid Contiguity and irreconcilable nonstandard asymptotics of statistical tests, by Pranab Kumar Sen and Antonio C. Pedroso-de-Lima September . 2011 IMS Bulletin . 17
Terence’s Stuff: Speaking, reading, writing
Terry Speed knows drift away if I’m not. fun and profit, and read mediocre or poor the importance of Few of us are naturally gifted speak- writing only when absolutely necessary. good communication. ers or writers. Most have to work hard Joseph Conrad, Jane Austen, G. H. Hardy He is inspired by a beautiful little book at it, and keep doing so. I’ll assume that and Marcia Angell all write magnificently. by R.D. Richardson: on applies to you, as it does to me and almost Articles in the New York Review of Books Emerson, on writing everyone else I know. How do we improve are invariably extremely well written, as are our speaking and writing? One answer is by those in Granta. Several of our professional n our business, communication is taking formal college or university courses, colleagues wrote beautifully: Florence important. It is no exaggeration to for example, on language and literature, Nightingale David, William Feller, and say that in our careers we sink, tread public speaking, technical or creative writ- David Freedman; and the mathematical Iwater or swim according to our ability to ing. There we can get instruction, practice and statistical biographer Constance Reid. communicate. Suppose that you are giving a and feedback. I always encourage people for Apologies here to speakers of other lan- presentation, and I am in your audience. At whom it is feasible to consider this option, guages for my focus on English. I can men- the start, you have my complete attention. that is, to seek professional help. Another tion Jacques Neveu or Paul-André Meyer I come expecting and hoping for a good approach is to dive into the “how to” books as French masters of writing, but that’s my experience, to learn something from you, on speaking and writing, though I can’t limit in other languages. There is a lot of and to enjoy myself. If I understand the comment on the value of this strategy. great writing out there, mostly by profes- words you use, the sentences you form, and We can also help ourselves. There are sional writers, of course, but they should be the chains of reasoning you construct, and two aspects to my self-help approach. our role models. Read and emulate them. if I can read and follow the material you One is like the way to get to Carnegie Like Emerson, I see the reading of good display, there is a good chance I’ll stay with Hall—practice, practice, practice—and the writing as key to improving our speaking you, perhaps to the end. If I struggle with other is reading. Ralph Waldo Emerson and writing. It helps us form what we want any of these aspects of your presentation, believed that first we read, then we write [the to say or write. Next comes the practice. I’ll expect extra value for the extra effort. title of the Richardson book on Emerson] Suppose you have a good first draft of your Otherwise, I’ll probably tune out. and gave the following writing tip: “Read talk or paper. You now need feedback: Communication is a two-way process for five hours a day.” Most of us are unlikely how good is it, and in what ways can it be most of the time, but not when you’re in to have our days as free for reading as his improved? Here is where you must rely on front, and I’m at the back. The onus is on were, so for us the question is: what should colleagues, friends or mentors, unselfish you to lure me, hook me, draw me in, and we read? Life is short, so my answer here people who are willing to spend their time eventually to land me. Actors, politicians, is unequivocal. Read only good writing for helping you improve your speaking or salespersons and evangelists all know this; writing. Such people aren’t as hard to find so should we. as you might think, but you should be Suppose that you have written a paper prepared to return the favour. Once started, and I am one of your potential readers. you then iterate the cycle of draft, receive Your title catches my attention. “This looks feedback, revise. It’s always hard, but there interesting,” I think, and I read the abstract. can be great satisfaction at the end: a talk Is this paper worth reading carefully, I well delivered, a paper flying through the wonder, as I read through the introduction. review process. Let’s give Emerson the last Do you get—and keep—my attention, as word: Happy is he … who writes from the you might if we were face to face? That love of imparting certain thoughts … who depends, both on my need to read your writes always to the unknown friend. paper, and on your writing. Again it is your words, sentences, and chains of reasoning “The first rule of writing is not to omit the thing you meant to say.” Sage advice from Ralph Waldo Emerson [left], that matter, and again, I’ll stay with you who may still be able to teach us a thing or two about if I’m learning and enjoying myself… and communication. 18 . IMS Bulletin Volume 40 . Issue 6
IMS meetings around the world IMS Annual Meetings, 2012 & 2014 IMS sponsored meeting IMS sponsored At a glance: 2012 World Congress/IMS Annual Meeting meeting July 9–14, 2012 2014 IMS Annual forthcoming Grand Cevahir Hotel & Convention Center, Istanbul, Turkey Meeting IMS Annual w http://www.worldcong2012.org/ July 7–11, 2014 The eighth World Congress in Probability and Statistics will be held in Istanbul from Sydney, Australia Meeting and July 9 to 14, 2012. It is jointly organized by the Bernoulli Society and the Institute of w TBC JSM dates Mathematical Statistics. Scheduled every four years, this meeting is a major worldwide The location for the event for statistics and probability, covering all its branches, including theoretical, method- 2014 IMS Annual ological, applied and computational statistics and probability, and stochastic processes. It Meeting has been 2012 features the latest scientific developments in these fields. selected as Sydney, IMS Annual Meeting Contacts: Elvan Ceyhan and Mine Çağlar, Co-chairs of the Local Organizing Australia. Details @ World Congress: Committee; Arnoldo Frigessi, Chair of the Program Committee. will follow, but İstanbul, Turkey, you can mark your July 9–14, 2012 calendars now! w http://www. worldcong2012.org/
JSM: San Diego, CA, July 28– August 2, 2012 Istanbul’s Bosphorus Bridge connects Europe (on the left) and Asia (right) w http://amstat. org/meetings/ Joint Statistical Meetings, 2012–2015 jsm/2012/ IMS sponsored meeting IMS sponsored meeting 2013 2012 Joint Statistical Meetings IMS Annual Meeting @ 2013 JSM IMS Annual Meeting July 28 – August 2, 2012 August 3–8, 2013 @ JSM: Montréal, San Diego, CA Montréal, Quebec, Canada Canada, August w http://amstat.org/meetings/jsm/2012/ w http://amstat.org/meetings/jsm/ 3–8, 2013 IMS Invited Program: Hans Mueller, University of California, Davis e [email protected]; IMS Contributed Program: Fang IMS sponsored meeting 2014 Yao, University of Toronto e [email protected] 2014 Joint Statistical Meetings Key dates August 2–7, 2014 IMS Annual Meeting: September 7: Invited session proposal submission deadline Boston, Mass., USA Sydney, Australia, September 30: CE proposal deadline w http://amstat.org/meetings/jsm/ July 7–11, 2014 December 21: Invited Program online JSM: Boston, MA, January 13: CTW proposal deadline IMS sponsored meeting August 2–7, 2014 February 1: Deadline for submission of abstracts for IOLs, Invited IMS Annual Meeting @ 2015 JSM posters, Topic-Contributed and Regular Contributed abstracts, and August 8–13, 2015 2015 Roundtables Seattle, Washington, USA May 10: Draft manuscript deadline w http://amstat.org/meetings/jsm/ IMS Annual Meeting @ JSM: Seattle, WA, August 8–13, 2015 September . 2011 IMS Bulletin . 19
ENAR, 2012–2014 IMS co-sponsored meeting The Second IMS Asia Pacific Rim Meeting IMS sponsored meeting July 1–4, 2012 2012 ENAR/IMS Spring Meeting Tsukuba, Japan April 1–4, 2012 w http://www.ims-aprm2012.org/ Washington DC, USA Program Chairs: Byeong U. Park e [email protected]), Runze Li e [email protected] w http://www.enar.org/meetings.cfm Meeting postponed from 2011 Since the massive earthquake struck Japan on March 11, the local organizing commit- IMS sponsored meeting tee and the scientific program committee decided to postpone the meeting until next 2013 ENAR/IMS Spring Meeting year. We have rescheduled it to July 1–4, 2012, and moved it to Tsukuba, the science March 10–13, 2013 city and academic center of Japan, which is about 60km from Tokyo. Orlando, Florida, USA We hereby cordially invite you all to attend the meeting next year, when we are w http://www.enar.org/meetings.cfm certain that you will witness a strong recovery of Japan from one of the most severe natural disasters in recent history. IMS sponsored meeting Akimichi Takemura, LOC Chair; Byeong Park & Runze Li, SC Co-Chairs 2014 ENAR/IMS Spring Meeting March 16–19, 2014 Baltimore, Maryland, USA IMS co-sponsored meeting w http://www.enar.org/meetings.cfm Colloquium in honor of Hans Rudolf Künsch on the occasion of his 60th birthday October 3–4, 2011, ETH Zurich, Switzerland IMS Reps: Peter Bühlmann, Marloes Maathuis, Sara van de Geer IMS co-sponsored meeting w https://stat.ethz.ch/events/Colloquium_Kuensch International Symposium in Statistics (ISS) Keynote speakers are Jim Berger (Duke University), Stuart Geman on Longitudinal Data Analysis Subject (Brown University), Peter Green (University of Bristol). Invited to Outliers, Measurement Errors, and/or speakers are: Rainer Dahlhaus (University of Heidelberg), Arnoldo Missing Values Frigessi (University of Oslo), Reinhard Furrer (University of July 16–18, 2012 Zurich), Havard Rue (Norwegian Univ. S&T Trondheim), Reto Memorial University, St. John’s, Canada Knutti (ETH Zurich), Christian P. Robert (Université Paris- w www.iss-2012-stjohns.ca Hans R. Künsch Dauphine). IMS Rep: Brajendra Sutradhar
IMS co-sponsored meeting IMS co-sponsored meeting International Statistics Conference 2011 International Conference Ars Conjectandi 1713–2013 December 28–30, 2011, Colombo, Sri Lanka October 15–16, 2013 w TBC Basel, Switzerland IMS Rep: Peter Hall. Organized by the Applied Statistics w http://www.statoo.ch/bernoulli13/ Association of Sri Lanka (ASASL). The meeting location is at the 2013 marks the 300th anniversary of the publication of Jacob water’s edge in the capital city of Sri Lanka. Bernoulli’s book, Ars Conjectandi, in 1713. A meeting has been organized to celebrate this: the “International Conference Ars IMS co-sponsored meeting Conjectandi 1713–2013” will be held October 15–16, 2013, in 36th Conference on Stochastic Processes and their Applications Basel, Switzerland. July 29 – August 2, 2013 IMS Representatives on the program committee are Hans University of Colorado, Boulder, USA Künsch and Lutz Dümbgen. w TBC 20 . IMS Bulletin Volume 40 . Issue 6
Other meetings around the world
PIMS-Mprime Summer School in Probability NEW Contemporary Issues and Applications of Statistics NEW June 4–29, 2012 January 2–4, 2012 University of British Columbia, Vancouver, Canada Kolkata, India w http://www.math.ubc.ca/Links/ssprob12/ w http://www.isical.ac.in/~cias This sixth school will feature two introductory courses, by Gregory The visionary Prof. Prasanta Chandra Mahalanobis planted the seed Miermont on planar maps and by Omer Angel on interacting of statistics in India. He realized the importance of developing the particle systems. Additionally, there will be a number of invited subject and founded the Indian Statistical Institute. As one of the lectures and presentations by participants. pioneers of policy-making after independence, he demonstrated how applications of statistics can help solve problems across a wide MCQMC 2012 NEW range of disciplines. He realized the need for interdisciplinary February 13–17, 2012 research where applications of statistics played a pivotal role. This Sydney, Australia endeavour often opened up many theoretical questions and led w http://www.mcqmc2012.unsw.edu.au/ to new ideas, results and even new directions for further research. The MCQMC Conference is a biennial meeting devoted to the Honouring the vision of this legendary scientist, Indian Statistical study of Monte Carlo (MC) and quasi-Monte Carlo (QMC) meth- Institute is organizing this conference on “Contemporary Issues and ods, the relationships between the two classes of methods, and their Applications of Statistics” (CIAS2012), to bring together statisti- effective application in different areas. cians and other scientists engaged in inter-disciplinary research involving applications of statistics from different parts of the world, International Conference on Trends and Perspectives in Linear NEW and provide a platform for them to interact among themselves and Statistical Inference [LinStat 2012], and the 21st International to share their thoughts on recent developments in different areas of Workshop on Matrices and Statistics [IWMS 2012] the subject. The success of the conference will definitely infuse fur- July 16–20, 2012 ther enthusiasm in statistical activities in India, particularly among Będlewo (near Poznań), Poland young researchers. w http://linstat2012.au.poznan.pl/ Contact: Saurabh Bhattacharya e [email protected] LinStat 2012 is the follow-up to the 2008 and 2010 editions held in Będlewo, Poland and in Tomar, Portugal; and IWMS 2012 is the SAMSI 2012–2013 Research Programs NEW 21st workshop in the sequence. The purpose of the meeting is to bring together researchers shar- Nonlocal Continuum Models (Summer Program) ing an interest in a variety of aspects of statistics and its applications July 17–20, 2012 as well as matrix analysis and its applications to statistics, and offer them a possibility to discuss current developments in these subjects. Computational Advertising (Summer Program) The conference will mainly focus on a number of topics. The August 6–17, 2012 topics that have been selected so far include estimation, prediction and testing in linear models, robustness of relevant statistical meth- Data-Driven Decisions in Healthcare (Year-Long Program) ods, estimation of variance components appearing in linear models, Opening Workshop August 26–29, 2012 generalizations to nonlinear models, design and analysis of experi- ments, including optimality and comparison of linear experiments, Statistical and Computational Methodology for Massive Data Sets and applications of matrix methods in statistics. (Year-Long Program) The work of young scientists is highly appreciated. The list of Opening Workshop September 9–12, 2012 Invited Speakers is opened by the winners of the Young Scientists Awards of LinStat’2010. The Scientific Committee will award the Opportunities to participate: visiting researchers, postdoctoral fel- best presentation and best poster. The awarded will be Invited lowships, graduate student fellowships and early career researchers. Speakers at the next edition of LinStat. Please see the advertisement on page 22. For more information and to apply, visit w www.samsi.info September . 2011 IMS Bulletin . 21
Employment Opportunities around the world
Australia: Parkville, Victoria United States: Washington DC The University of Melbourne, Dept of Mathematics and Statistics Georgetown University, Department of Biostatistics Lecturer (US equivalent: Assistant Professor) Assistant Professor in Biostatistics/Bioinformatics http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8385648 http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8407272 United States: Washington DC Canada: Waterloo, ON U.S. Census Bureau University of Waterloo Associate Director for Demographic Programs Actuarial Science – Tenure-Track http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8415657 http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8516540 United States: West Lafayette, IN Canada: Waterloo, ON Purdue University, Department of Statistics University of Waterloo Assistant Professor Biostatistics – Tenure-Track http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8545969 http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8516498 United States: Williamstown, MA Canada: Waterloo, ON Williams College University of Waterloo Assistant Professor of Statistics Statistics – Tenure-Track http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8397181 http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8516413 United States: Brunswick, ME France: Rennes Bowdoin College, Department of Mathematics ENSAI, National School for Statistics & Information Analysis Statistics Position Research Fellow Biostatistics http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8436857 http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8515810 United States: Ann Arbor, MI United States: Los Angeles, CA University of Michigan, Department of Statistics UCLA Department of Mathematics Assistant Professor Faculty Positions 2012-2013 http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8341842 http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8407075 United States: Columbia, MO United States: Stanford, CA University of Missouri, Department of Statistics Stanford University Assistant Professor Faculty Opening http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8332793 http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8516768 United States: Chapel Hill, NC United States: Stanford, CA UNC Department of Biostatistics Stanford University Post-doctoral Fellow Stein Fellow http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8502143 http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8432752 United States: Seattle, WA United States: Stanford, CA Fred Hutchinson Cancer Research Center Stanford University Staff Scientist Assistant Professor/Associate Professor http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8415680 http://jobs.imstat.org/c/job.cfm?site_id=1847&jb=8432735
::: Search our online database of the latest jobs around the world for free at http://jobs.imstat.org ::: 22 . IMS Bulletin Volume 40 . Issue 6
Employment Opportunities around the world
Hong Kong: Kowloon United States: Ithaca, New York The Hong Kong University of Science and Technology, Department of Mathematics Cornell University Dept of Mathematics Faculty Position(s) The Department of Mathematics at Cornell The Department of Mathematics invites applications for tenure-track faculty position(s) at University invites applications for two H.C. the rank of Assistant Professor in all areas of mathematics. Other things being equal, prefer- Wang Assistant Professors, non-renewable, ence will be given to areas consistent with the Department’s strategic planning. 3-year position beginning July 1, 2012. Applicants should have a PhD degree and strong experience in research and teaching. Successful candidates are expected to pursue Applicants with exceptionally strong qualifications and experience in research and teaching independent research at Cornell and teach may be considered for position(s) above the Assistant Professor rank. three courses per year. A PhD in mathemat- Starting rank and salary will depend on qualifications and experience. Fringe benefits ics is required. The Department actively include medical/dental benefits and annual leave. Housing will be provided where appli- encourages applications from women and cable. Initial appointment will be on a three-year contract, renewal subject to mutual agree- minority candidates. Applicants must apply ment. A gratuity will be payable upon successful completion of the contract. electronically at http://www.mathjobs.org. Applications received on or before 31 December 2011 will be given full consideration For information about our positions and for appointment in 2012. Applications received afterwards will be considered subject application instructions, see http://www. to the availability of position(s). Applicants should send their curriculum vitae together math.cornell.edu/Positions/positions.html. with at least three research references and one teaching reference to the Human Resources Applicants will be automatically considered Office, HKUST, Clear Water Bay, Kowloon, Hong Kong. Applicants for position(s) above for all eligible positions. the Assistant Professor rank should send curriculum vitae and the names of at least three Deadline December 1, 2011. Early research referees to the Human Resources Office. More information about the University applications will be regarded favorably. is available on the University’s homepage at http://www.ust.hk. (Information provided by Cornell University is an Affirmative Action/ applicants will be used for recruitment and other employment-related purposes.) Equal Opportunity Employer and Educator.
United States: Research Triangle Park, NC Announcing the SAMSI 2012 - 2013 Research Programs Nonlocal Continuum Models (Summer Program) July 17 - 20, 2012 Computational Advertising (Summer Program) August 6 - 17, 2012 Data-Driven Decisions in Healthcare (Year-Long Program) Opening Workshop August 26 - 29, 2012 Statistical and Computational Methodology for Massive Data Sets (Year-Long Program) Opening Workshop September 9 - 12, 2012 Opportunities to Participate: •Visiting Researchers •Postdoctoral Fellowships •Graduate Student Fellowships •Early Career Researchers For more information and to apply, visit www.samsi.info
::: Advertise current job opportunities for only $250 for 60 days ::: See http://jobs.imstat.org for details ::: September . 2011 IMS Bulletin . 23
USA: Atlanta, GA Georgia Tech The School of Mathematics at Georgia Tech is accepting applica- tions for faculty positions at all ranks and in all areas of pure and applied mathematics and statistics. Applications by highly qualified can- didates from groups underrepresented in the mathematical sciences are particu- larly encouraged. See www.math.gatech. edu/resources/ employment for more details and application instruc- tions.
United States: Ithaca, New York Cornell University, Department of Mathematics The Department of Mathematics at Cornell University invites applications for two tenure-track Assistant Professor positions, or higher rank, pending administrative approval, starting July 1, 2012. The searches are open to all areas of Mathematics with an emphasis on the areas of probability, number theory, and PDE. The Department actively encourages applications from women and minority candidates. Applicants must apply electronically at http://www.mathjobs.org. For information about our positions and application instructions, see: http://www.math.cornell.edu/Positions/positions.html. Applicants will be automatically considered for all eligible positions. Deadline November 1, 2011. Early applications will be regarded favorably. Cornell University is an Affirmative Action/Equal Opportunity Employer and Educator.
::: Search our online database of the latest jobs around the world for free at http://jobs.imstat.org ::: 24 . IMS Bulletin Volume 40 . Issue 6 International Calendar of Statistical Events IMS meetings are highlighted in maroon with the logo, and new or updated entries have the NEW or UPDATED symbol. t means telephone, f fax, e email and w website. Please submit your meeting details and any corrections to Elyse Gustafson at [email protected]
September 2011 October 23–27: Pucón, Chile. Third Latin American Meeting on Bayesian Statistics (COBAL) and XXXVIII Chilean National September 5–9: Lisbon, Portugal. 17th European Young Meeting of Statistics (JNE) w http://cobal2011.usach.cl Statisticians Meeting w http://www.fct.unl.pt/17eysm October 24–26: Knoxville, TN, USA. NIMBioS Investigative September 7–8: Statistical Center of Statistics Korea, Daejeon, Workshop: Mathematical Modeling of Intracellular Movements South Korea. Third International Workshop on Internet Survey w http://nimbios.org/workshops/WS_intracellular_mv.html Methods w www.kostat.go.kr October 28–30: Prague, Czech Republic. Analytical Methods in September 12–16: Mathematical Biosciences Institute, Columbus, Statistics (AMISTAT) w http://amistat2011.karlin.mff.cuni.cz Ohio. Workshop on New Questions in Probability Theory Arising in Biological Systems w http://mbi.osu.edu/2011/ws1description.html November 2011 September 12 – December 16: Institute for Pure and November 7–9: Łódź, Poland. Multivariate Statistical Analysis Applied Mathematics, Los Angeles, USA. Mathematical and Conference w http://www.msa.uni.lodz.pl Computational Approaches in High-Throughput Genomics w www.ipam.ucla.edu/programs/gen2011/ November 25–27: Lahore, Pakistan. 3rd International Conference on Statistical Sciences w http://www.icss3.co.nr/ September 13–16: Jaca, Spain. Statistics, Probability and Operations Research (SPO 2011) w http://metodosestadisticos. unizar.es/~jaca2011 December 2011
September 24: Cambridge, MA. 2011 New England Symposium December 18–21: Amman, Jordan. 11th Islamic Countries on Statistics in Sports w http://www.amstat.org/chapters/boston/ Conference on Statistical Sciences (ICCS-11) w http://www.iccs11. nessis11.html isoss.net/
September 25–28: Ribno, Bled, Slovenia. Applied Statistics 2011 w December 28–30: Colombo, Sri Lanka. International http://conferences.nib.si/AS2011 Statistics Conference 2011. w TBC
September 26–27: Birmingham, Alabama. Statistical Analyses for December 28–31: Hong Kong, China. International Conference Next Generation Sequencing w http://www.soph.uab.edu/ssg/ on Advances in Probability and Statistics Theory and Applica- courses/nhgri_r13/ngsstat tions: A celebration of N. Balakrishnan’s 30 years of contributions to statistics. e [email protected] w http://faculty.smu.edu/ngh/ October 2011 icaps2011.html
October 3–4: ETH Zurich, Switzerland. Colloquium in January 2012 honor of Hans Rudolf Künsch on the occasion of his 60th birth- day w https://stat.ethz.ch/events/Colloquium_Kuensch January 1–6: Hyderabad, India. 22nd Annual Conference of The International Environmetrics Society w www.ties2012.com/ October 12–14: Washington DC/ Silver Spring MD. Conference on Risk Assessment and Evaluation of Predictions w http://brac. NEW yJanuar 2–4: Kolkata, India. Contemporary Issues and umd.edu/~Risk2011/Main.htm Applications of Statistics (CIAS2012) w http://www.isical.ac.in/~cias October 18–20: Harvard Medical School, Cambridge, Mass. 2011 Non-clinical Biostatistics Conference w http://www.hsph.harvard. edu/ncb2011/ September . 2011 IMS Bulletin . 25
January 23–27: Centre International de Rencontres Mathématiques March 2012 (CIRM), Marseille, France. Concentration inequalities and their applications w http://www.cirm.univ-mrs.fr/ March 14–16: Hong Kong. IAENG International Conference on Data Mining and Applications 2012 w www.iaeng.org/IMECS2012/ ICDMA2012.html February 2012 March 30–31: Washington DC. Information and Econometrics NEW February 13–17: MCQMC 2012 w http:// Sydney, Australia. of Networks w www.american.edu/cas/economics/info-metrics/ www.mcqmc2012.unsw.edu.au/ workshop/workshop-2012-spring.cfm
April 2012 June 2012 April 1–4: Washington DC, USA. 2012 ENAR/IMS Spring June 3–6: Guelph, Ontario, Canada. SSC Annual Meeting w TBC Meetings. w http://www.enar.org/meetings.cfm NEW June 4–29: University of British Columbia, Vancouver, April 18–20: Poznań, Poland. International Congress of Polish Canada. PIMS-Mprime Summer School in Probability w http:// Statistics to celebrate the 100th anniversary of the Polish Statistical www.math.ubc.ca/Links/ssprob12/ Association w http://www.stat.gov.pl/pts/ June 18–22: MIT, Cambridge, MA, USA. The 2012 Stochastic Networks Conference w http://stoch-nets-2012.lids.mit.edu/
June 23–26: Boston, MA, USA. ICSA 2012 Applied Statistics Symposium. w TBC Meeting organizer’s to do list June 25–29: Kyoto, Japan. 2012 ISBA World Meeting w http://www2.e.u-tokyo.ac.jp/~isba2012/
July 2012
July 1–4: Tsukuba, Japan. IMS Asia Pacific Rim Meetings. w http://www.ims-aprm2012.org/ (meeting postponed from July 2011 due to the earthquake)
July 3–6: University of Oslo, Norway. Third biennial International Statistical Ecology Conference w http://www.cees.uio.no/news/2010/isec2012.html
July 9–14: Istanbul, Turkey. IMS Annual Meeting 2012 in conjunction with 8th World Congress in Probability and Statistics. w http://www.worldcong2012.org/
Continues on page 26
26 . IMS Bulletin Volume 40 . Issue 6
International Calendar continued
July 2012 continued August 2013
July 16–18: Memorial University, St. John’s, Canada. Inter- August 3–8: Montréal, Canada. IMS Annual Meeting at national Symposium in Statistics (ISS) on Longitudinal Data JSM2013. w http://amstat.org/meetings/jsm/ Analysis Subject to Outliers, Measurement Errors, and/or Missing August 24–31: Hong Kong. International Statistical Institute: 59th Values w www.iss-2012-stjohns.ca ISI World Statistics Congress w www.isi2013.hk NEW July 16–20: Będlewo (near Poznań), Poland. International Conference on Trends and Perspectives in Linear Statistical Infer- October 2013 ence [LinStat 2012], and 21st International Workshop on Matri- ces and Statistics [IWMS 2012] w http://linstat2012.au.poznan.pl/ October 15–16: Basel, Switzerland. International Conference Ars Conjectandi 1713–2013 w http://www.statoo.ch/bernoulli13/ NEW July 17–20: SAMSI, NC, USA. Nonlocal Continuum Mod- els [SAMSI Research Program] w www.samsi.info March 2014 July 28 – August 2: San Diego, California. JSM2012. w http://amstat.org/meetings/jsm/2012/index.cfm March 16–19: Baltimore, Maryland. 2014 ENAR/IMS Spring Meeting. w http://www.enar.org/meetings.cfm August 2012 July 2014 NEW August 6–17: SAMSI, NC, USA. Computational Advertis- ing [SAMSI Research Program] w www.samsi.info July 7–11: Sydney, Australia. 2014 IMS Annual Meeting. w TBC NEW August 26–29: SAMSI, NC, USA. Data-Driven Decisions in Healthcare [SAMSI Research Program] Opening Workshop w www.samsi.info August 2014 August 2–7: Boston, MA. JSM2014 and ASA’s 175th September 2012 Anniversary. w TBC
NEW September 9–12: SAMSI, NC, USA. Statistical and Com- putational Methodology for Massive Data Sets [SAMSI Research August 2015 Program] Opening Workshop w www.samsi.info August 8–13: Seattle, WA. JSM2015. w TBC
March 2013
March 10–13: Orlando, Florida. 2013 ENAR/IMS Spring Are we missing something? If you know of Meeting. w http://www.enar.org/meetings.cfm any statistics or probability meetings which aren’t listed here, please let us know. Email July 2013 the details to Elyse Gustafson at erg@imstat. org. We’ll list them here in the July 29 – August 2: University of Colorado, Boulder, USA. 36th Conference on Stochastic Processes and their Applications w Bulletin, and online too, at TBC www.imstat.org/meetings September . 2011 IMS Bulletin . 27
Membership and Subscription Information Information for Advertisers Journals General information: The IMS Bulletin and webpages are the official The scientific journals of the Institute of Mathematical Statistics are news organs of the Institute of Mathematical Statistics. The IMS The Annals of Statistics, The Annals of Probability, The Annals of Applied Statistics, The Annals of Applied Probability, and Statistical Science. The Bulletin, established in 1972, is published 8 times per year. Print IMS Bulletin is the news organ of the Institute. circulation is around 2,000 paper copies, and it is also free online http://bulletin.imstat.org Individual and Organizational Memberships in PDF format at , posted online about two Each individual member receives the IMS Bulletin (print and/or weeks before mailout (average downloads over 8,000). Subscription electronic) and may elect to receive one or more of the five scientific to the IMS Bulletin costs $90. To subscribe, call 877-557-4674 journals. Members pay annual dues of $103. An additional $53 is (US toll-free) or +1 216 295 2340 (international), or email staff@ added to the dues of members for each scientific journal selected. imstat.org. http://imstat.org Reduced membership dues are available to full-time students, new The IMS website, , established in 1996, graduates, permanent residents of countries designated by the IMS receives over 30,000 visits per month. Public access is free. Council, and retired members. Organizational memberships are available to departments, corporations, government agencies and Advertising job vacancies other similar research institutions at $155 per year. A single 60-day online job posting costs just $250.00. We will also Individual and General Subscriptions include the basic information about your job ad (position title, Subscriptions are available on a calendar-year basis. Individual subscriptions are for the personal use of the subscriber and must location, company name, job function and a link to the full ad) in be in the name of, paid directly by, and mailed to an individual. the IMS Bulletin at no extra charge. See http://jobs.imstat.org Individual subscriptions for 2011 are available to The Annals of Applied Probability ($166), The Annals of Applied Statistics ($166), The Advertising meetings, workshops and conferences Annals of Probability ($166), The Annals of Statistics ($166), Statistical Science ($146), and IMS Bulletin ($113). General subscriptions Meeting announcements in the Bulletin and on the IMS website at are for libraries, institutions, and any multiple-readership use. http://imstat.org/meetings are free. Send them to Elyse Gustafson; General subscriptions for 2011 are available to The Annals of Applied see http://www.imstat.org/program/prog_announce.htm Probability ($360), The Annals of Applied Statistics ($320), The Annals of Probability ($360), The Annals of Statistics ($390), Statistical Science ($210), and IMS Bulletin ($95). Airmail rates for delivery outside Rates and requirements for display advertising North America are $109 per title. Display advertising allows for placement of camera-ready ads for IMS Bulletin journals, books, software, etc. A camera-ready ad should be sent as The IMS Bulletin publishes articles and news of interest to IMS a grayscale PDF/EPS with all fonts embedded. Email your advert to members and to statisticians and probabilists in general, as well as Audrey Weiss, IMS Advertising Coordinator [email protected] or details of IMS meetings and an international calendar of statistical see http://bulletin.imstat.org/advertise events. Views and opinions in editorials and articles are not to be understood as official expressions of the Institute’s policy unless so Dimensions: width x height Rate stated; publication does not necessarily imply endorsement in any 1/3 page 4.93" x 4" (125.2 x 102 mm) $215 way of the opinions expressed therein, and the IMS Bulletin and its 1/2 page 7.5" x 4" (190 x 102 mm) $270 publisher do not accept any responsibility for them. The IMS Bulletin 2/3 page 4.93" x 8" (125.2 x 203 mm) $325 is copyrighted and authors of individual articles may be asked to sign Full page (to edge, a copyright transfer to the IMS before publication. 8.75" x 11.25" (222 mm x 285.8 mm) $380 including 1/8" bleed) The IMS Bulletin (ISSN 1544-1881) is published eight times per year Full page (within usual 7.5" x 9.42" (190 mm x 239.3 mm) $380 in January/February, March, April/May, June/July, August, September, Bulletin margins) October/November and December, by the Institute of Mathematical Statistics, 3163 Somerset Dr, Cleveland, Ohio 44122, USA. Deadlines and Mail Dates for IMS Bulletin Periodicals postage paid at Cleveland, Ohio, and at additional mailing offices. Postmaster: Send address changes to Institute of Mathematical Issue Deadline Online by Mailed Statistics, 9650 Rockville Pike, Suite L3503A, Bethesda, MD 20814- 1: January/February December 1 December 15 January 1 3998. 2: March February 1 February 15 March 1 3: April/May March 15 April 1 April 15 Copyright © 2011 by the Institute of Mathematical Statistics. 4: June/July May 1 May 15 June 1 Printed by The Sheridan Press, 450 Fame Avenue, Hanover, PA 5: August July 1 July 15 August 1 17331, USA. 6: September August 15 September 1 September 15 7: Oct/Nov September 15 October 1 October 15 8: December November 1 November 15 December 1 Volume 40 . Issue 6 the The purpose of the Institute is to foster the nextissue is development and dissemination October/ of the theory and applications of November statistics and probability 2011 IMS: Organized September 12, 1935
ISSN (0091-1798) http://projecteuclid.org/aop Read it online at AOP July 2011 the new website: THE ANNALS bulletin.imstat.org of PROBABILITY
AN OFFICIAL JOURNAL OF THE DEADLINES INSTITUTEOF MATHEMATICAL STATISTICS submissionsfor
September 15, Articles then November 1 The TASEP speed process ...... GIDEON AMIR,OMER ANGELAND BENEDEK VALKÓ 1205 Loop-erased random walk and Poisson kernel on planar graphs ARIEL YADINAND AMIR YEHUDAYOFF 1243 Please see inside On monochromatic arm exponents for 2D critical percolation VINCENT BEFFARA AND PIERRE NOLIN 1286 the back cover for Some stochastic process without birth, linked to the mean curvature flow KOLÉHÈ A.COULIBALY-PASQUIER 1305 subscription details Markov processes on time-like graphs ...... KRZYSZTOF BURDZYAND SOUMIK PAL 1332 Reconstruction for the Potts model ...... ALLAN SLY 1365 and information for Lack of strong completeness for stochastic flows XUE-MEI LIAND MICHAEL SCHEUTZOW 1407 advertisers, including Backward stochastic dynamics on a filtered probability space GECHUN LIANG,TERRY LYONSAND ZHONGMIN QIAN 1422 all our deadlines and Harnack inequality for SDE with multiplicative noise and extension to Neumann semigroup on nonconvex manifolds...... FENG-YU WANG 1449 requirements The regularizing effects of resetting in a particle system for the Burgers equation GAUTAM IYERAND ALEXEI NOVIKOV 1468 A probabilistic approach to Dirichlet problems of semilinear elliptic PDEs with singular coefficients ...... TUSHENG ZHANG 1502 Concentration of the information in data with log-concave distributions SERGEY BOBKOVAND MOKSHAY MADIMAN 1528 Journal Spectrum of large random reversible Markov chains: Heavy-tailed weights on the complete graph . . . CHARLES BORDENAVE,PIETRO CAPUTOAND DJALIL CHAFAÏ 1544 Approximating the moments of marginals of high-dimensional distributions alerts ROMAN VERSHYNIN 1591 For alerts and special Convergence of joint moments for independent random patterned matrices ARUP BOSE,RAJAT SUBHRA HAZRAAND KOUSHIK SAHA 1607 information on all the IMS journals, sign up at the IMS Lists site http://lists.imstat.org Vol. 39, No. 4—July 2011