Math & Stats

The PDF version of the March 2012 IMS Bulletin is online now: click the “latest issue” link in the bar above.

The eighth World Congress in Probability and Statistics, which includes the 2012 IMS Annual Meeting, will be held July 9–14, 2012, in Istanbul, Turkey (see www.worldcong2012.org).

The Congress organisers are pleased to announce that four new sources of funding have been confirmed, from:
1. ISI/World Bank: for residents of developing countries
2. NSF: for US-based participants
3. Elsevier/Stochastic Processes and Their Applications: for young researchers
4. IMU: for residents of developing countries

In addition, as previously announced, IMS members who have childcare responsibilities can apply for funds to cover up to 80% of the costs of childcare while they attend the Congress. Details on the Childcare Initiative are at http://www.imstat.org/meetings/childcare.htm. The deadline for the Laha Travel Awards has passed.

More information on these funding opportunities is posted at http://www.worldcong2012.org/?p=financial_support.

The deadline for abstract submission for the Congress is March 16, 2012 (with notification of acceptance by April 1, 2012); the early registration deadline is May 21, 2012.
Istanbul has a wealth of cultural attractions, including the Blue Mosque, pictured below, from Hagia Sofia.
Görüşürüz! (See you later!)

Photo: Damien Roué

US National Academy of Engineering elects members
The United States National Academy of Engineering has announced its new members for 2012. Congratulations to IMS members Peter W. Glynn (Thomas W. Ford Professor and chair, management science and engineering department, Stanford University) and Michael S. Waterman (University Professor, USC Associates Chair in Natural Sciences, and professor of biological sciences, computer science, and mathematics, College of Letters, Arts, and Sciences, University of Southern California, Los Angeles). Peter was selected for “contributions to simulation methodology and stochastic modeling,” and Michael for “development of computational methods for DNA and protein sequence analyses.” The full list of members is at http://www8.nationalacademies.org/onpinews/newsitem.aspx?RecordID=02092012

American Association for the Advancement of Science Fellows
The American Association for the Advancement of Science (AAAS) has elected 539 Fellows for their contributions to science and technology. Among them were several IMS Fellows and/or members.
In the Section on Mathematics were Ingrid Daubechies, Duke University, and Claudia Neuhauser, University of Minnesota Rochester.
In the Section on Statistics were: George Casella, University of Florida; Dipak K. Dey, University of Connecticut; Wing Kam Fung, University of Hong Kong; Sastry G. Pantula, National Science Foundation; and Xiaotong Shen, University of Minnesota.

Elected members of ISI
The International Statistical Institute elected 17 new members in the third round of 2011, among whom were five IMS members: Jiti Gao, Stefano Iacus, Ranjan Maitra, Axel Munk, and Haiyan Wang. The list is at http://isi-web.org/membership/membership-elections-2011-round-3.

Spiegelman Award
The 2011 Mortimer Spiegelman Award was presented in November to Sudipto Banerjee, University of Minnesota. The award, given annually since 1970, is presented to an outstanding public health statistician under the age of 40. Sudipto Banerjee was recognized for contributions to spatial and environmental statistics. The committee also acknowledged his leadership and strong success in external funding, PhD supervision, short course instruction, software development, and journal editing and refereeing.
See the call for nominations for the 2012 award here.

New IMS Program Secretary
IMS Council has approved the appointment of Judith Rousseau as IMS Program Secretary for a term of July 12, 2012 to August 10, 2015. She will replace Guenther Walther on the Executive Committee, following his four years of service. Judith is a professor at Ceremade at the Université Paris Dauphine: see her webpage at http://www.ceremade.dauphine.fr/~rousseau/.

New Editors for Annals of Statistics

Peter Hall and Runze Li will serve as Editors of The Annals of Statistics for the three-year term January 1, 2013 to December 31, 2015.
Peter, an IMS Fellow, has served IMS in many capacities, including IMS President, Council, and Committees on Fellows, on Nominations and on Special Lectures. Runze is also an IMS Fellow, and received the 2012 United Nations’ World Meteorological Organization Gerbier-Mumm International Award. The current Annals of Statistics Editors (for the period 2010 to 2012) are Peter Bühlmann and T. Tony Cai. The homepage of the journal is http://www.imstat.org/aos/ and IMS members can read all papers, past and present, at http://projecteuclid.org/aos/

New Editor for Annals of Applied Probability

Timo Seppäläinen will be the next Editor of The Annals of Applied Probability. Timo is a full professor in the Mathematics Department at the University of Wisconsin, Madison. He is an IMS Fellow, and has served as Associate Editor for the Annals of Probability and Annals of Applied Probability, and most recently, until December 2011, served as the Editor of Electronic Communications in Probability (see below).

New home for Electronic Journal of Probability and Electronic Communications in Probability

The Electronic Journal of Probability (EJP) and Electronic Communications in Probability (ECP) are electronic journals, jointly sponsored by the IMS and the Bernoulli Society. These journals have a new online home.

Since their inception in 1995, the content of EJP and ECP has been freely available to individuals. Until recently, the published versions of the journals were hosted on servers at the Department of Mathematics, University of Washington. Due to computing environment changes at UW and editorial management needs associated with continued growth of the journals, the contents have recently migrated to new websites at http://ejp.ejpecp.org/ and http://ecp.ejpecp.org/ (the old websites automatically forward to these sites).

The servers for these sites are maintained by the Public Knowledge Project (PKP) at Simon Fraser University in Canada. PKP is also the provider of the open source software (OJS) used for editorial management and publishing of the journals. The IMS and the Bernoulli Society are sharing the modest costs.

We encourage you to visit the new websites, to consider submitting an article to EJP or ECP, and ask you to please consider making a donation to the open access fund in support of our open access journals.

The Editors of the IMS-supported journal Brazilian Journal of Probability and Statistics issue a call for papers for a forthcoming special issue:

We are pleased to announce a special issue of the Brazilian Journal of Probability and Statistics. The issue is to be titled, “Innovative Methodology in Modern Applications.” The idea is to present original contributed work and specially invited papers from leading researchers on statistical methodology developed for modern areas of application. The guest editors are Pranab Kumar Sen (UNC at Chapel Hill), Pedro A. Morettin (University of São Paulo), Gisela Tunes da Silva (University of São Paulo) and Aluísio Pinheiro (University of Campinas).

High dimensional data models are especially prominent in modern applications in interdisciplinary research. The targeted areas include: time series models in functional domain, multivariate methods, especially in interdisciplinary research, resampling plans and data mining for nonstandard data models, bioinformatics and environmetric problems, and other contemporary problems arising in modern applications.

Manuscripts should be submitted on-line to the Brazilian Journal of Probability and Statistics via http://imstat.org/bjps/mansub.html no later than September 1, 2012. Please make sure to choose “Special issue” as Manuscript Type. The submission should contain a cover letter indicating that the manuscript is being submitted to the Innovative Methodology in Modern Applications issue. A review and revision process will follow. Accepted manuscripts should be revised no later than February 28, 2013. Upon final acceptance papers will be available on-line. Authors should follow the instructions available at http://imstat.org/bjps/mansub.html for manuscript preparation formatting and reviewing standards. For expedient publication of accepted papers the use of the template found at http://www.e-publications.org/ims/support/bjps-instructions.html is highly recommended.

The Statistics Section of the American Public Health Association (APHA) invites nominations for the 2012 Mortimer Spiegelman Award honoring a statistician, 40 years of age or younger, who has made outstanding contributions to health statistics, especially public health statistics.

The award was established in 1970 and is presented annually at the APHA meeting. The award serves three purposes: to honor the outstanding achievements of both the recipient and Spiegelman, to encourage further involvement in public health of the finest young statisticians and to increase awareness of APHA and the Statistics Section in the academic statistical community. More details about the award including the list of the past recipients and more information about the Statistics Section of APHA may be found at http://sites.google.com/site/aphastatistics/benefits/spiegelman.

To be eligible for the 2012 Spiegelman Award, a candidate must have been born in 1972 or later. Please send electronic versions of nominating letter and the candidate’s CV to the 2012 Spiegelman Award Committee Chair, Rafael A. Irizarry (rafa@jhu.edu). Please state in the nominating letter the candidate’s birthday. The nominator should include one to two paragraphs in the nominating letter that describe how the nominee’s contributions relate to public health concerns. A maximum of three supporting letters per nomination can be provided.

Nominations for the 2012 Award must be submitted by April 1, 2012.

The title of this column (including the two extra periods) is that of Emanuel Derman’s book, which has the eye-catching subtitle, “Why Confusing Illusion with Reality Can Lead to Disaster, on Wall Street and in Life.” The book has six chapters grouped into three pairs:
I. Models;
II. Models Behaving, and
III. Cohomology of Simplicial Complexes.
Just kidding, of course, your guess is right. One thing that I am confident that you cannot guess based on the title, is that the first chapter is concerned with the author growing up in South Africa as “the accidentally-conceived last child of Jewish parents who immigrated from Poland (now Belarus) to Cape Town in the mid-1930s.” Ten interesting pages are devoted to his membership in Habonim, a coeducational Zionist youth movement, which, he says “left its marks on me, many of them good.”

Large parts of the book are about physics, for examples the puzzles of the positron, which “Dirac found to be a hole in the sea of electrons.” Reading this material made me happy that there aren’t such paradoxical things in probability to make one think deeply about the nature of reality and turn to thinkers like Schopenhauer and Spinoza for help. See the complex network of pleasure, pain and desire on pages 86–87. Most of the book is considerably more fun and filled with interesting quips. Speaking about gravity, Derman says, “Newton was confident that the power of the distance was precisely 2. Had he been a social scientist he would probably have proposed a power of 2.05 ± 0.31.”

Some of Derman’s stories are borrowed from others: Freeman Dyson wrote in his essay Frogs and Birds, “Some mathematicians are birds, others are frogs. Birds fly high in the air and survey broad vistas of mathematics out to the far horizons. They delight in concepts that unify our thinking … Frogs live in the mud below and see only the flowers that grow nearby. They delight in the details of particular objects, and they solve problems one at a time.”

Derman began his professional life as a physicist, “studying fundamentals and mastering theory. Then in 1985, I migrated to the center of the quant world at Goldman Sachs. My colleagues were as smart as academics but more interesting. Their work was an interdisciplinary mix of modeling, mathematics, statistics, and programming.” Academic experts on finance may think they know what mathematics he is referring to, but when the narrative turns to an event in his academic life at Columbia, we learn that he had to use Google to figure out what “the fundamental theorem of finance was.” I have read the statement he includes on page 141 and can’t figure it out. Google must have grown up since then.

The lack of one familiar fact is more than made up for by the new ones that I saw there. If you haven’t seen Fischer Black’s paper on “Noise” in the Journal of Finance, volume 41 (1986), 529–543, then you should go to JSTOR and read it. Derman devotes a lot of ink to discussion of deficiencies of the efficient market model and CAPM, but perhaps the most persuasive argument is given in the figure on page 184 that compares simulated stock prices to four years of the S&P 500. I almost said that the picture is worth the price of the book, but the picture can be had for the price of one photocopy so that doesn’t make much sense economically.

Of course, at its heart this is a book about models. Derman explains that the widespread shock at the failure of quantitative models in the mortgage crisis of 2007 results from a misunderstanding of the difference between models and theories. Theories describe and deal with the world on its own terms and must stand on their own two feet. Models stand on someone else’s feet. Models try to squeeze the blooming buzzing confusion into a miniature Joseph Cornell box, and then, if it more or less fits, assume that the box is the world itself. Intuition is more comprehensive. It unifies the subject with the object, the understander with the understood, and the archer with the bow.

People expect either too much or too little from financial models. One must begin boldly but expect little. If a little success is actually attained, one must grow greedy. Then when one has gone only a little too far, desist. Derman has more concrete advice, quoting from a Modeler’s Hippocratic Oath that he wrote with Paul Wilmott. The third point is “I will not be overly impressed by mathematics. I will never sacrifice reality for elegance without explaining to its end users why I have done so.”

Wise words and there many of them in the book that I haven’t quoted. At $26 it is a bargain, and I am looking forward to reading his other book, “My Life as a Quant: Reflections on Physics and Finance.”

Johannes H.B. Kemperman’s research and teaching career of 50 years included ten years at Purdue, 25 years at the University of Rochester, and ten years at Rutgers. During this time he produced three major books, over 100 publications, and 23 PhD students. He was a Fellow of IMS and of the American Association for the Advancement of Sciences, and a Correspondent of the Royal Dutch Academy of Sciences. He served in editorial posts at the Annals of Mathematical Statistics, Annals of Probability, Annals of Statistics, Stochastic Processes and their Applications, and Aequationes Mathematicae. His wide research interests led to outstanding work in number theory, group theory, analysis including asymptotic expansions, functional equations, mathematical biology, probability, and statistics.

Johannes, or less formally, Joop (pronounced “Johp” in Dutch) was born in Amsterdam in 1924, the eldest of five children in a middle-class family. His mathematical ability led to his getting scholarships at the University of Amsterdam, where he received his BS in 1945, and PhD in 1950 in Mathematics/Physics. During World War II, he strongly opposed the Nazis, who occupied the Netherlands. He spent much of the war hiding in haylofts, reading mathematics.

After passing the PhD-qualifying “Doctoraal Examen” in 1948, he worked for three years full-time as a research associate in the Department of Applied Mathematics at the prestigious Mathematische Centrum in Amsterdam. Occasionally the Department would receive requests for help on applied problems, and Joop solved several groundwater flow problems posed by the Dutch waterworks. Most of the time Joop was free to work on his own research interests in number theory, pure analysis, probability and statistics. Joop got his PhD in Amsterdam, The Netherlands, on December 20, 1950. This was the beginning of an outstanding research career.

In 1951 Joop went on a Fulbright grant to the Mathematics Department at Purdue University, where he was subsequently offered a tenure track Assistant Professorship and became a Full Professor in 1959. In June 1953 Joop married Wilna Ypma from The Netherlands, whom he had known for five years, and after a brief honeymoon they returned to Purdue. They later had five children together: Steve, Bruce, Hubert, Ingrid and Eric.

At Purdue, probability and statistics were taught by the Math Department faculty. This teaching sparked Joop’s interest to the extent that about half of his career publications were in probability and statistics. A notable example was Joop’s 1956 Annals of Mathematical Statistics paper on generalized tolerance limits, that was stimulated by teaching a course in nonparametric statistics. Joop also began work on his classical monograph The First Passage Problem for a Stationary Markov Chain which he completed in 1959. Joop also continued his research in number theory and analysis.

In 1961 Joop was recruited to the Mathematics Department at the University of Rochester, later named as Fayerwether Professor. At the time there was not a separate statistics department. In 1963, Allen Wallis, a strong promoter of statistics departments, became Chancellor of the University of Rochester. In 1968 Jack Hall was recruited as chair of a new statistics department, and attracted excellent people. The Department, with a total of about six full-time equivalent lines, used partial lines for joint appointments, including Joop whose tenure remained in mathematics.

At Rochester, Joop was active in stimulating research in theoretical biology, mathematical economics, mathematical physics and operations research, in addition to his active program of teaching and research in mathematics and statistics. He was an active participant in developing policy and leadership in both of these departments.

In 1985 Joop retired from Rochester and joined the Statistics Department at Rutgers, and was also a full voting member of the Mathematics Department. Joop found Rutgers an exciting mathematical environment with stimulating talks and speakers. He did joint work with Art Cohen and Harold Sackrowitz in addition to his wide-ranging research.

A Rutgers Statistics seminar in 1992 by Joel Cohen stimulated a collaboration that led to a joint book that deals with many different ways to compare information channels. The 1998 book Comparisons of Stochastic Matrices with Applications in Information Theory, Statistics, Economics, and Population Sciences, jointly with Joel Cohen and Gheorghe Zbaganu, is breathtaking in its perspectives, and was awarded the Gheorghe Lazăr Prize of the Romanian Academy in 2000. Joel notes how “Joop contributed many innovative and unexpected ideas and techniques. Joop’s creativity, combined with his broad mathematical background, came to the fore in the startling proofs…”

Joop was known for his willingness to listen and contribute to a great variety of mathematical problems. This trait is illustrated by his large number of co-authors, and the large number of mathematicians who cite his results. Once he became interested in a subject he often would investigate it in great detail. This shows itself in the large number of quite long papers of Joop’s, often with results spread out over years; see, for instance, his papers on information theory, or on distribution mod 1, or on moment problems.

An important contribution to probability theory, motivated by questions in statistics, and in particular to sequential analysis, deals with the exit time and exit place of a random walk Sn from a strip parallel to the time-axis. More generally Joop allows Sn to be a Markov chain. Only rarely are explicit calculations possible, but Joop derived interesting and powerful methods to attack such problems. In coding theory, Joop found simple proofs and generalized basic results.

Some of the papers Joop was most proud of were in number theory, related group theory, and functional equations. Examples are his 1957 paper in the Transactions of the American Mathematical Society on a generalized functional equation, and his 1960 Acta Math paper dealing with small sumsets in Abelian groups. Throughout his career Joop was interested in the use of probability theory in number theory. In particular one finds a number of papers which apply probability to ‘distribution mod 1’.

Joop passed away on June 13, 2011, surrounded by family. Joop’s beloved Wilna had passed away in 1995. Joop is survived by four of his children—Bruce, Hubert, Ingrid and Eric—and three daughters-in-law, one son-in-law and eight grandchildren.

Joop was an exceptionally nice person and a true gentleman. He was a wonderful colleague and friend. We are all grateful that we had him with us for many years. He leaves a beautiful personal and professional legacy that will be cherished by all who knew him.

Written by Jack Hall and Joe Naus

Some of this material is summarized from: Naus, J.I., “A conversation with Johannes H.B. Kemperman” Statistical Science 15 (November 2000) 396–408.

The Statistical and Applied Mathematical Sciences Institute (SAMSI) has teamed up with the Institute for Computational and Experimental Research in Mathematics (ICERM) and several institutes in India (CMI, IISc,ISI, IMSC) to form the Virtual Institute for Mathematical and Statistical Sciences (VI-MSS). The Virtual Institute will sponsor joint workshops, research visitors and graduate educational activities with support from the US National Science Foundation. VI-MSS is part of a larger NSF initiative for international collaborations known as Science Across Virtual Institutes.

As part of this initiative, SAMSI announces three new activities:

1. A workshop on Environmental Statistics will be held at SAMSI on June 18–20, 2012.
There will be invited speakers from both the US and India. Registration for the meeting will shortly be open and will include the opportunity to participate in a poster session. Major themes of the workshop include: measuring and monitoring the environment; health effects of air and water pollution; and climate change.

2. A workshop on Topics in Probability will be held December 18–20 2012, at the Chennai Mathematical Institute (CMI).
This workshop will consist of 12 to 15 one-hour lectures spanning a broad range of topics of current interest in probability. We expect that approximately half of the speakers will be from US universities and the other half from universities in India.

3. In addition to the above, we invite applications from US participants to take part in research visits to India.

Research visits may be up to six months in duration for the purpose of extended research collaboration with a faculty member at one of the participating Indian institutes. To be considered for a research visit during the 2012–2013 academic year, applications should be received at SAMSI no later than March 31, 2012. Applications may be made through the SAMSI website www.samsi.info and will be coordinated with ICERM.

Further queries about any of these activities may be sent to Richard Smith, Director of SAMSI at rls@samsi.info

Letter to the Editor

Dear Editor

My name is Czesław Stępniak. I have been a member of the IMS for over 20 years. I have written more than 50 articles to 31 mathematics and statistics journals. I would like to share some of my experience in the use Elsevier’s leading citation database Scopus.

Recently, by browsing the database, I found my name in only 14 documents. To my surprise my last work (J. Statist. Plann. Inference, 141, 2489–2493) in this database is assigned to “Steogonekpniak” and another one (Metrika, 36, 291–298), which has 20 citations, is assigned to “Stepolhkepniak.” Of course none of these works can be reached by the usual route through “author search” because neither Stepolhkepniak and Steogonekpniak exist as authors. What is more, Scopus has deleted all documents related to my previous employment (covering more than 25 years).

In my opinion the matter is serious because this database is the main information source for all authors and reviewers, who would be firmly convinced that information contained therein is accurate. In this way all errors are repeatedly reproduced in the published articles. I encourage other IMS members to discover new secrets of the citation database Scopus.

I suspect that my experience is only the tip of the iceberg.

Czesław Stępniak
University of Rzeszów, Poland

Editor’s note: Dr. Stępniak has, of course, reported his concerns to Scopus. Get in touch if you have had similar experiences: leave a comment below, or email bulletin@imstat.org

The Editor’s decision is final; we may edit your letter before publication; publication does not necessarily imply endorsement of the opinions expressed therein, and the IMS Bulletin and its publisher do not accept responsibility for them.

Rajshahi University Statistics Department held its Golden Jubilee Celebrations on December 23–24, 2011, to commemorate the fiftieth anniversary of the founding of the department. The Statistics Department at Rajshahi University, Bangladesh, was founded by the late Professor Khondkar Manwar Hossain in 1961. It was through his dedicated efforts that the department now has about four hundred undergraduate majors, Master’s, and PhD students every year, with over 30 teaching faculty members. The department has a strong research group under the mentorship of Professor M. Nasser, Chairman of the Statistics Department.

The two-day Golden Jubilee Celebrations drew over 1600 participants which included among others alumni and current students and also former and present teachers of the statistics department. On the opening day there was a joy rally at the heart of the campus where national, university and two other flags were ceremoniously hoisted and doves were released by the guests of honor and then the participants walked through the campus in a procession to the assembly hall.

At the opening session Dr. Mir Masoom Ali, George and Frances Ball Distinguished Professor of Statistics Emeritus at Ball State University, USA, delivered the Golden Jubilee Keynote Lecture, followed by the lectures by Pro Vice Chancellor and Vice Chancellor of Rajshahi University and the Chairman of the Statistics Department. The rest of the afternoon was spent in viewing the poster sessions. The posters included many timely and important topics and were mainly prepared by the graduate students under the mentorship of their professors. The second day was mainly spent with invited plenary session presentations. The Golden Jubilee festivities ended in the evening with a fine cultural show that included songs and dances by local and national artists. The whole event was an outstanding success.

The 11th biennial Islamic Countries Conference on Statistical Sciences (ICCS-11) has elected Professor Ali S. Hadi of The American University of Cairo (AUC), Egypt as the new President of the Islamic Counties Society of Statistical Sciences (ISOSS). He became the President-elect in December 2009. He takes over the role of President from Professor Shahjahan Khan, University of Southern Queensland, Australia who held the position from 2005 to 2011. Ali is currently a Distinguished University Professor and a former Vice Provost at AUC. He was also a former Head of the Department of Social Statistics at Cornell University, USA, where he is currently a Stephen H. Weiss Presidential Fellow and Professor Emeritus.

The ICCS conference was held at the University of Management and Technology (UMT), Lahore, Pakistan, from 19–22 December, 2011. The conference attracted delegates from Australia, Africa, South East Asia, Middle East, Europe, and North America. Unfortunately, due to the current situation in Pakistan many of our colleagues from the West who registered for the conference could not participate in it. The Higher Education Commission of Pakistan provided free accommodation for a large number of international and out of city participants in its newly built Guest House in Gulberg III, Lahore, Pakistan.

The theme of the conference was “Statistics for Strategies of Development.” The speakers in the Opening Session discussed different aspects of the application of statistics in the modern world. In his talk, the Rector of the UMT, Professor Hassan Sohaib Murad, focussed on the dominant role of statistics in business and industry. He also emphasized the need of statistics for business executives and the inclusion of courses on quantitative methods at the undergraduate level. The founding President of ISOSS and Rector of National College of Business Administration and Economics (NCBA&E), Professor Munir Ahmed, gave a brief genesis of ISOSS and explained the need for unity among scientists to enhance the harmonious development of knowledge for the benefit of humanity. Professor Ali Hadi expressed appreciation towards the ISOSS team in Lahore for their excellent job organising an international conference at very short notice. In his speech, Professor Shahjahan Khan, outgoing President of ISOSS explained the stages of development of statistics and its wide range of applications in almost all areas of research, modern science, business and every walk of life. He also thanked everyone who helped build ISOSS House in Lahore, and those who organised four international conferences during his term. The Vice-President of ISOSS, Professor Mohammad Hanif Mian, appreciated support from everyone, especially the sponsors, UMT and NCBA&E, and participants from home and abroad.

The inaugural keynote address was presented by Ali Hadi. He entertained the audience through showing graphs to explain relationships among a large number of variables on the parallel coordinate while speaking on “Multi-Class Data Exploration Using Space Transformed Visualization Plots.” Other keynote addresses included “Statistical Meta-analysis of Ordinal Categorical Data” by Shahjahan Khan (Australia), “Universally Optimal Designs in Blocks of Size Three” by Munir Akhtar (Pakistan), “A Brief Review of Geoinformatics Analysis on Poverty Data in Indonesia” by Asep Saefuddin (Indonesia), and “Improved Estimation in Simple Linear Regression Model with Autocorrelated Errors” by Bashir Khan (Canada).

In the Business Session the proposal for the creation of a Centre of Excellence in Statistical Research was approved, to train young statisticians of the OIC member states. Participants agreed on Kuala Lumpur as the venue of the proposed Centre which may pull together the available resources of the University of Malaya, University Putra Malaysia and University Kebangsaan Malaysia. Any affiliated national statistical societies/associations of the OIC member states will be recognized as ISOSS chapters. An International Board of Directors will be formed involving regional representatives to enhance the future growth and activities of ISOSS.

The new President of ISOSS, Ali Hadi emphasized the importance of increasing the membership of ISOSS, strengthening its activities, stability, and sustainability through regular sponsorship and increased collaboration with organisations with similar objectives and professional goals. The Gala Dinner of the conference was held at ISOSS House. Many participants were very impressed to see the ISOSS headquarters for the first time. The names of the people who donated funds for the construction were displayed at the entrance of the newly-built office. The building was completed early 2011.

An interesting day tour to one of the largest salt mines in the world was organised on the last day of the conference. Khewar Salt Mine is about three hours’ drive from Lahore (towards Islamabad). It was discovered in 326 BC. The 17-storey mine has an area of 110 square km—all inside an ordinary-looking mountain. The main visiting locations are two kilometres inside the mountain, accessed by a special rail train. The miners have been extracting salt from the mine for the last 300 years. The participants on the tour were very happy with the arrangements and the exhibits in and around the mine.

It was announced that the next ISOSS conference (ICCS-12) will be hosted by Qatar University in Doha, Qatar, from 19–22 December, 2012. It may be noted that originally the ISOSS conferences were held in the even years but due to unavoidable reasons the last few conferences could not be held in even years. However, from 2012 ISOSS plans to organize its biennial conferences in even years in order to alternate with the International Statistical Institute’s conferences in odd years.

A post-conference workshop on “Successfully publishing research articles in international journals” was presented by Professor Shahjahan Khan at the NCBA&E on 23 December 2011. Eighty academics and researchers including PhD and MPhil students participated in the day-long event. The workshop covered various aspects of preparation of research articles, selection of appropriate journal, formatting and referencing of articles, online submission, review and revision process, citation of published articles, impact factor of journals, international collaboration and co-authors, and H and G indices.

Reported by Shahjahan Khan, University of Southern Queensland, Australia; and Ali S Hadi, The American University of Cairo, Egypt.

March 1, 2012 is the deadline for requests for NIMBioS support for Working Groups, Investigative Workshops, Postdoctoral Fellows, Sabbaticals, and Short-term Visitors for activities beginning summer/fall 2012. All areas of research at the interface of biology and mathematics will be considered, but we are especially interested in activities expanding beyond the areas of research supported to date. Potential organizers of activities in areas of molecular biology, cell biology, network biology, immunology and systems biology are particularly encouraged to submit requests for support of Working Groups or Investigative Workshops. NIMBioS, located at the University of Tennessee-Knoxville, is an NSF-sponsored initiative to foster interdisciplinary research at the interface between mathematical and biological sciences. The institute’s mission is to cultivate cross-disciplinary approaches in mathematical biology and to develop a cadre of researchers who address fundamental and applied biological problems in creative ways. Other NIMBIoS sponsors include DHS and USDA, with additional support from the University of Tennessee-Knoxville.
More details are posted at http://www.nimbios.org
Related Links:
NIMBioS Working Groups: http://www.nimbios.org/workinggroups/
NIMBioS Investigative Workshops: http://www.nimbios.org/workshops/
Postdoctoral Fellowships: http://www.nimbios.org/postdocs/
Sabbaticals: http://www.nimbios.org/visitors/sabbaticalShort-term

Hira L. Koul, President of the Indian Statistical Association, writes: The Indian Statistical Association was established in 1960 at Bombay (now Mumbai). The late Professor M.C. Chakrabarty, the founding Head of the Department of Statistics, University of Bombay, was its first Secretary and Prof. A. M. Kshirsagar was the first President. Chakrabarty played a key role in establishing the Association and starting and running its journal, the Journal of the Indian Statistical Association (JISA: http://stats.unipune.ernet.in/jisa.html). Besides publishing this journal, the Association also organizes and supports conferences, workshops, and meetings in India. The current president and secretary of the Association are Hira L. Koul and Uttara V. Naik-Nimbalkar, respectively. A more detailed account of the Association can be found in the following articles: Adke, S.R. (2010). An early history of ISA and JISA, JISA, 48, 181–188; Chakrabarty, M.C. (2010). My pilgrimage to Bombay, JISA, 48, 171–178 (written some time in 1960–61); and Kshirsagar, A.M. (1972). Professor M.C. Chakrabarty, JISA, 10, 1–8.

The first issue of JISA was published in January, 1963, with Chakrabarty being the Chief Editor of the Journal until his premature death in 1972. Subsequently, the late V.S. Huzurbazar and S.R. Adke were appointed as the Chief Editor and Secretary in 1972, respectively. Professor Adke streamlined the journal during 1972–1980 and also helped to create a sound financial corpus for the Association. Naik-Nimbalkar is the current Chief Editor of the JISA.

This year JISA is completing its fifty years. To celebrate the event, a special Golden Jubilee volume of the journal containing the following papers is being published.
1. N. Balakrishna and A. J. Lawrance. Development of product autoregressive model.
2. A. Bose, R. S. Hazra and K. Saha. Extremum of circulant type matrices: a survey.
3. A. Chatterjee and P. Hall. High dimensional classification when useful information comes from many, perhaps all features.
4. Y.P. Chaubey, J. Li, A. Sen and P. K. Sen A new smooth density estimator for non-negative random variables
5. R. Dutta, M. Bogdan and J. K Ghosh Model selection and multiple testing – A Bayes and empirical Bayes overview and some new results.
6. A. K. Jammalamadaka and S. R. Jammalamadaka A growth curve approach to analyzing multiple-valued expression data.
7. S.N. Lahiri, X.L. Nguyen, J. Yang, Z. Zhu, and P. Banerjee. Wireless sensor networks: Statistical issues and challenges
8. S.H. Ong and A. SenGupta. Bivariate and multivariate circular distributions by mixtures.
9. B. L. S. P. Rao. Test for dominance between mean functions for two-sample functional data.
10. G. A. Satten and S. Datta. Minimum distance type estimation of transformation parameters.
11. A. Schick and W. Wefelmeyer. Convergence in weighted L1-norms of convolution estimators for the response density in nonparametric regression.
12. W. Stute and T. L. Anh. Principal component analysis of martingale residuals
13. T. Umashanger, T. N. Sriram, and J. Lee. Simultaneous robust estimation in finite mixtures: The continuous case.
14. P. Vellaisamy Simpsons paradox and collapsibility

Have you ever calculated a Pearson correlation without doing a scatter plot? Long ago I vowed never to do so, but from time to time I forget my vow. I did so a couple of weeks ago when a student and I wanted to use our data to see if we could replicate a plot we’d seen in a published paper. Our plot showed a noisy but more or less monotonically decreasing relationship between the Pearson correlation of lagged pairs of proportions, and the lag. It went from about 0.9 at lag 1 down to about 0.3 at lag 1,000 and 0.2 at lag 1,500, where it levelled out. As we saw this in six different data sets, and it was in broad agreement with the published plots, we accepted it as replication and felt happy. However, we had another approach to measuring the decay of that association with lag, and it dropped monotonically to independence at lags around 100. This reduced our level of happiness. After some thought I remembered my vow, and looked at the scatter plots other than at lag 1, which we had examined. By lag 100, the scatter plot looked pretty bad, but gave a Pearson correlation of 0.7, as there were lots of points near (0,0) and (1,1). By lag 200, there was not much evidence of meaningful correlation, with the points mostly on the edges of the square connecting (0,0), (0,1), (1,1) and (1,0). We concluded that Pearson’s linear correlation was a very bad measure to be using in this context. By lag 100, the correlations were meaningless. (It might have been our data: correlations are quite susceptible to influence by selection bias.)

This little experience resonated with two themes I’ve been reading about recently, quite independently of the project involving lagged correlations. One was the strong aversion Tukey and others had to calculating correlation coefficients, something of which I should have been aware long ago. C. P. Winsor started the Society for the Suppression of the Correlation Coefficient whose guiding principle was “that most correlation coefficients should never be calculated.” Tukey was a member, one who frequently held “that correlation coefficients are justified in two and only two circumstances, when they are regression coefficients, or when the measurement of one or both variables on a determinate scale is hopeless.” It’s not clear how many other members of this society there were, but Fisher agreed with Tukey and Winsor, writing over a decade earlier that, “regression coefficients are of interest and scientific importance in many classes of data where the correlation coefficient, if used at all, is an artificial concept of no real utility.”

Of course Pearson’s r is for linear association, it is not robust, correlation is not causation, and in many—perhaps most—contexts, regression coefficients are more meaningful than correlations. But what if we are genuinely interested in the association between two variables, and not at all in the regression of one on the other? For example, when the variables are the same, separated in space or time by some lag, even if the measurement is on a determinate scale? Our interest in the association between certain proportions along the genome was exactly of this kind. I’d go further and say that we expected it to be caused by something, and were seeking some indication of the spatial scale over which this cause might be operating? We might even be interested in periodicity in the lagged association measure, and be led to take the Fourier transform of what we calculated. One of the papers we were following did just that. Should we be discouraged from calculating lagged correlation coefficients, or some other measure of association? I certainly hope not. But we must look at our scatter plots!

What are our options for measuring association, if it’s not linear? Over the years a large number of these have appeared, including measures using ranks (Spearman’s ρ, Kendall’s τ), nonparametric measures (Hoeffding’s D, following Mosteller, Blomqvist’s q’), mutual information (following Shannon, Linfoot’s r1, and many others), principal curve-based measures (Delicado & Smrekar’s covGC), and more recently, distance and Brownian correlation (Székeley & Rizzo’s dCor and CorW).

In parallel with these new ideas, several authors (including Fisher, Maung, Gebelein and Rényi) going back to Hirschfeld have used maximal linear correlation over nonlinear transformations of the original variables. Very recently, a new measure of association building on mutual information (Reshef et al’s MIC) was proposed. The authors claimed that it gave a meaningful measure for a wide range of nonlinear relationships (almost) independent of the nature of the nonlinearity.

In 1954 Tukey asked, “Does anyone know when the correlation coefficient is useful, as opposed to when it is used? What substitutes are better for which purposes?”

Can we answer him yet?

The eighth World Congress in Probability and Statistics) will be held on July 9–14, 2012, in Istanbul, Turkey. Scheduled every four years, it is jointly organized by IMS and the Bernoulli Society. This meeting is a major worldwide event for statistics and probability, covering all its branches including theoretical, methodological, applied and computational statistics and probability, and stochastic processes. It features the latest scientific developments in these fields.

The congress will present recent developments and the state-of-the-art in a variety of modern research topics, with in-depth sessions on applications of these disciplines to other sciences, industrial innovation and society. It will feature several special plenary lectures: eight named lectures, five medallion lectures, and a public lecture, presented by leading specialists. In addition, there will be 40 invited sessions highlighting topics of current research interests, as well as a large number of contributed sessions and posters.

IMS President Ruth Williams provided information on the IMS initiatives available to help members attend the congress in the October/November Bulletin. These include the IMS Laha Travel Awards and Childcare Initiative, as well as the possibility of a US funding agency to help support junior researchers from US institutions to attend the Congress. More information on these and other funding opportunities is posted at http://www.worldcong2012.org/?p=financial_support.

The venue of the meeting is Grand Cevahir Hotel & Convention Center which has many conference amenities. The host city, Istanbul is more than 8,000 years old, and is a multi-cultural, easy to access and cosmopolitan city that bridges Europe and Asia. It has a unique cultural fusion of east and west, offering such famous cultural attractions as Hagia Sophia, Sultanahmet, Topkapı Palace and Maiden’s Tower. Istanbul offers memorable historical tours and easy city transportation (including ground, underground, and sea transportation). For shopping, it has both old-style colorful bazaars and modern malls.

Furthermore, Istanbul is Europe’s third-largest city (and the world’s 21st), yet it is one of the safest cities, having the lowest crime rate among European capitals (research conducted by the EU Crime and Security Consortium). The host country, Turkey is a must-see place, famous for its hospitality, centrally-located, and with a developing economy and a young, vibrant population.

Abstract submission and registration are open now. The deadline for abstract submission is March 16, 2012 (with notification of acceptance by April 1, 2012); the early registration deadline is May 21, 2012.

On behalf of the Program Committee and the Local Organizing Committee, we invite you to join us in Istanbul for this exciting event. With your participation, the 2012 World Congress will be a memorable meeting.

Elvan Ceyhan and Mine Çaglar, Co-chairs of the Local Organizing Committee, and Arnoldo Frigessi, Chair of the Program Committee

Itai Benjamini to be Inaugural Schramm Lecturer

As announced in the last issue, the IMS and the Bernoulli Society have recently cooperated to create a new joint lecture in probability and stochastic processes, named in honor of Oded Schramm (1961–2008). The lecture will be given annually and will be featured at meetings (co)-sponsored by the IMS or the Bernoulli Society with a strong attendance by researchers in probability and stochastic processes.

Professor Itai Benjamini, Renee and Jay Weiss Chair in the Department of Mathematics at Weizmann Institute, Israel, will deliver the inaugural Schramm Lecture. Itai, who collaborated closely over many years with Oded, said he was “very touched by the invitation.”

Itai’s Schramm Lecture will be delivered at the Stochastic Processes and their Applications meeting to be held in Boulder, Colorado, July 29 to August 2, 2013. See http://math.colorado.edu/spa2013/ for more information.

Symposium honours Johan Fellman

To honour Professor Johan Fellman on his 80th birthday, a symposium was arranged on November 25, 2011, at Hanken School of Economics, Helsinki, Finland. The program comprised: Professor Gunnar Rosenqvist (Hanken School of Economics, Finland): Introduction; Professor Kenneth Nordström (University of Oulu, Finland): Convexity of the Inverse and Moore-Penrose Inverse; and Professor Katarina Juselius (University of Copenhagen. Denmark): On the Role of Theory and Evidence in Macro Economics.

See http://www.hanken.fi/public/en/finance_satistics_seminars_and_workshops#document2.

New Society for Non-Parametric Statistics

The International Society for Non-Parametric Statistics (ISNPS) was founded in 2010 by three IMS Fellows, M.G. Akritas, S.N. Lahiri, and D.N. Politis, with a mission, “to foster the research and practice of nonparametric statistics, and to promote the dissemination of new developments in the field via conferences, books and journal publications.”

ISNPS has a distinguished Advisory Committee that includes R. Beran, P. Bickel, R. Carroll, D. Cook, P. Hall, W. Hardle, R. Johnson, B. Lindsay, E. Parzen, P. Robinson, M. Rosenblatt, G. Roussas, T. SubbaRao, and G. Wahba, as well as a Charting Committee consisting of over 50 prominent researchers from all over the world. The nature of ISNPS is uniquely global, and its international conferences are designed to facilitate the exchange of ideas and latest advances among researchers from all around the world in cooperation with established statistical societies such as IMS and the ISI.

The First Conference of ISNPS is scheduled to take place in Chalkidiki, Northern Greece, June 15–19, 2012, and is co-sponsored by IMS and ISI: see www.isnpstat.org for more details. IMS members who are interested in the new society and/or its first conference may contact ISNPS at isnps@stat.tamu.edu

New Managing Editor for EJP/ECP

There is a new Managing Editor for the joint IMS-Bernoulli Society journals Electronic Journal of Probability and Electronic Communications in Probability. Djalil Chafaï takes over from Philippe Carmona for three years. Papers can be accessed at http://www.math.washington.edu/~ejpecp/

NISS receives five-year grant from NSF for Triangle Census Research Network

The National Institute of Statistical Sciences (NISS) and Duke University have received a grant from the US National Science Foundation (NSF) and the US Census Bureau for the Triangle Census Research Network (TCRN). The award, one of eight nationwide under the NSF-Census Research Nodes program, is for nearly $3 million and covers a five-year period. Jerome Reiter (who is the Mrs. Alexander Hehmeyer Associate Professor of Statistical Science at Duke University) is principal investigator (PI) on the project; Alan Karr (director of NISS) is the co-PI.

The grant will be used to improve how federal statistical (“FedStats”) agencies disseminate data to the public and to researchers. Specifically, the TCRN will enhance FedStats agencies’ capabilities by developing broadly-applicable methodologies in three interrelated areas: (i) disseminating public use data with high utility and acceptable disclosure risk, (ii) handling missing data and correcting faulty data in large complex surveys, and (iii) integrating information from multiple data sources. The TCRN will also offer educational opportunities to postdoctoral fellows, graduate students, and statisticians at federal agencies, helping to train future leaders in data dissemination research and practice.

The FedStats agencies collect data of all kinds that affect many people, including the decennial census, unemployment numbers and the Consumer Price Index. NISS has been and is in collaboration with many of these agencies, including (in addition to the Census Bureau) the Bureau of Labor Statistics, the Bureau of Transportation Statistics, the Energy Information Administration, the National Agricultural Statistics Service, the National Center for Education Statistics and the National Center for Health Statistics. Among NISS’s achievements are methods used nationally to produce high school graduation rates and crop forecasts, as well as a plethora of techniques and tools that support dissemination of high quality information derived from confidential data.

By building on these achievements as well as creating new theory and methodology applicable to major Census Bureau data products, the TCRN’s research will improve the hundreds of secondary analyses of these datasets. The interdisciplinary team of the TCRN, which consists of statisticians, economists, political scientists and operations researchers, will use these data products to answer questions in aging, economics, and social welfare that have important implications for policy making.

“The TCRN will improve the way we handle missing and faulty data by integrating paradigms from statistics and operations research,” explained Karr, “The team will also develop nonparametric Bayesian approaches for multiple imputation of missing data in high dimensions with longitudinal and multi-level aspects, as well as address central issues in data integration.”

Festschrift for S. Rao Jammalamadaka

Professor S. Rao Jammalamadaka, an IMS Fellow, ISI Member and ASA Fellow, has been honored with Advances in Directional and Linear Statistics: A Festschrift for Sreenivasa Rao Jammalamadaka. This volume of papers, published by Springer-Verlag, consists of articles written by students, colleagues and collaborators from over 20 countries, and covers a variety of research topics of particular interest to him. See http://www.springer.com/statistics/statistical+theory+and+methods/book/978-3-7908-2627-2

Piled Higher and Deeper, by Jorge Cham. This cartoon, originally published at www.phdcomics.com, echoes the sentiments in Terry Speed’s last column. You can read what he has to say about looking—really looking—for solutions, in his column on page 13 of the Jan/Feb 2012 issue, or here.

When we heard about the new Museum of Mathematics (MoMath) that is being built in New York, a museum that bills itself as “an exciting project that will change the face of American education in mathematics,” we naturally wanted to find out how statistics and probability will be represented…

George Hart is Chief of Content at MoMath. He responded, “We are in the midst of designing exhibits for our Fall 2012 opening and we plan to touch on many areas of mathematics, including probability and statistics. One exhibit concept on our drawing board is related to the Galton board, an exhibit that is probably familiar to your members, in which falling balls hit pins to make a series of independent left and right moves until they collect at the bottom of the exhibit in an approximate Gaussian distribution. In the variation we would like to build, the visitor first draws a probability distribution on a touch screen, then the pins of the exhibit move slightly (under computer control) so that the falling balls now land with the given target distribution. We have some engineering work to make the pin mechanism sufficiently accurate, but it strikes us as an engaging exhibit idea with great potential for getting people to think about probability and statistics. I’d be interested to hear if your readers have ever seen such an adjustable Galton board elsewhere, and if they have suggestions for other exhibits we should include.” Email George at hart@momath.org.

Artist’s rendering of the proposed adjustable Galton Board

The Museum of Mathematics is a unique and innovative institution that strives to enhance public understanding and perception of mathematics as an evolving, creative, and aesthetic human endeavor. The Museum has secured space at the north end of Madison Square Park in Manhattan, and it will offer a wide variety of hands-on, interactive exhibits that will spark excitement and wonder in kids and adults of all ages. MoMath will be North America’s only museum devoted to the wonders of mathematics and its many connections.

While the museum’s space is being developed, they are already running several programs promoting math as interesting, exciting, and fun. MoMath’s popular traveling exhibition, the Math Midway, is in the midst of a national and international tour, and the monthly Math Encounters presentation series routinely draws hundreds. The streets of Manhattan provide the perfect setting for inspiring math-themed walking tours, and three of the city’s boroughs will play host to MoMath’s exciting new math tournaments in the spring.

See http://momath.org for more information. And you can help the Museum open its doors by going to momath.org/contribute

Photo of the exterior of MoMath

Charles Bernard Bell, Jr. was born in New Orleans on August 20, 1928. To the statistical community he was affectionately called “Chuck.” He graduated from Xavier University at an early age, and then attended Notre Dame University where he earned a masters degree in mathematics and statistics in 1948, and a doctorate in 1953. His graduation was coupled with his marriage to Mary Drye of Los Angeles, California. Chuck was the first African-American to receive both masters and doctorate degrees from Notre Dame.

He worked as a research engineer at the Douglas Aircraft Company, but was more interested in an academic position and joined the faculty at Xavier University for two years before visiting Stanford University for a year. From 1958 to 1966 Chuck was on the faculty of San Diego State University and again from 1981 to 1992, at which time he retired. During the period from 1966 to 1981 he was at Case Western University (1966–1968), the University of Michigan (1968–1971), Tulane University (1971–1977), and the University of Washington (1977–1981).

Chuck loved languages, and conquered many. He could speak, read, and write in Spanish, German, Russian, Dutch, French, Italian, and Swahili. I remember on one of his visits to Stanford that he would talk to someone in one language, and turn around to someone else and speak in another language. In fact, he sought out international visitors in order to practice his multi-lingual conversational skills. As an example of this linguistic prowess, he co-authored four papers in Spanish, one in German, a monograph in French and one in Dutch.

Chuck and his family traveled extensively: to the Mathematical Institute in Amsterdam (1964); to the University of Madrid (1964–1965); to the University of Vienna (1965); to the Institute of Statistics in the University of Paris (1965–1966); to the University of Erlangen (1966); to the University of Göttingen (1972); and to the University of Costa Rica (1986). He was a welcome visitor at Stanford, and visited during the summers of 1965, 1987, and 1991.

Throughout his life Chuck was interested and active in African-American affairs. He was Chair of the Minority Statisticians Committee of the American Statistical Association, and in 1968 worked with African mathematicians in Mombasa, Kenya. In 1969 he helped develop courses in mathematics for teachers in Nigeria, and in 1975, at Tulane, gave a workshop for black undergraduates.

In 1974 the Mathematical and Physical Sciences Directorate of the National Science Foundation created a program in applied mathematics and statistics, and Chuck served as its second Program Director for statistics during 1975–76.

Chuck had a distinguished research career. His thesis was on “structures of (probability) measure spaces.” His early statistical work was on various nonparametric problems, and this remained an intermittent focus of his work.

Distribution-free tests occupied his attention in several papers:

(i) structure of distribution-free statistics,

(ii) characterization of multi-sample distribution-free statistics,

(iii) “optimal” one-sample distribution-free tests,

(iv) distribution-free tests of independence,

(v) distribution-free tests of randomness.

A second general area of research was stochastic processes, e.g., uniform renewal processes, Weibull-type Poisson processes, and Pareto renewal processes. During his career he published 39 papers. Chuck enjoyed collaborations, and wrote papers with 25 colleagues, one being David Blackwell. Honors include election as a Fellow of the American Statistical Association, the Institute of Mathematical Statistics, and Fulbright Fellow.

It was noted in an obituary written for the Mathematical Association of America that the Bell and Thomas families (Thomas was his mother’s maiden name) had relations who were teachers going back 125 years, contributing a total of 400 years of service to that profession. Both Chuck and Mary, who teaches French, continued this tradition.

Chuck died on October 26, 2010, in Los Angeles. He was predeceased by his daughter, Elaine E. Bell, and is survived by his wife of more than 56 years, Mary Drye Bell, his children Karen Bell Shirley and her husband Jensen H. Shirley, Roslyn A. Bell, C.B. Bell, III, and his granddaughters Jessica A. Shirley and Ebony K. Shirley.

All who knew him will remember his infectious laugh, his love of gumbo, his upbeat attitude, and his friendship.

Written by Ingram Olkin, Stanford University

Photo courtesy of C. B. Bell III: http://cbbell.com