Home » Member News, Obituaries, People News

Obituaries for May 2019

1 May 2019 No Comment

Ayala (Aya) Cohen

Ayala (Aya) Cohen passed away February 20, 2019, at 79. She is survived by her husband, Elisha Cohen, their two sons, and seven grandchildren.

Cohen was an Israeli statistician at the Technion – Israel Institute of Technology’s Faculty of Industrial Engineering and Management and head of the Technion’s Statistics Laboratory. Born in Tel Aviv on January 27, 1940, to German-born parents, Cohen lost her father when she was 9 years old. In 1957, upon graduating from high school, she joined the Israel Defense Forces, where she served for two years as a mathematics/physics tester for pilot cadets. She earned her bachelor’s degree in mathematics and physics in 1962 from the Hebrew University in Jerusalem and her PhD in statistics from The Johns Hopkins University in 1967.

From 1967 to 1969, Cohen worked with the statistics group at AT&T Bell Labs, where she also spent two sabbaticals and several summers. She joined the Technion in 1970 and worked with researchers in multiple disciplines throughout her career. She published both methodological and applied articles, focusing on behavioral sciences and medicine. Her areas of expertise included applied multivariate analysis, hierarchical modeling, time series analysis, and biostatistics. She was past-president of the Israel Statistical Association and a member of the Advisory Committee for Statistical Methodology, nominated by the National Statistical Council.

In 1988, along with Paul Feigin, Cohen established the Technion Statistics Laboratory and later became its head, a position she held until her passing. For more than 30 years, the lab staff under her leadership provided statistical consulting to a wide range of clients and organizations, often developing and publishing methodological innovations. Her consulting style was highly professional but also personal and generated among her clients a large cohort of devoted friends for life. The statistics lab provided students of all disciplines with sound statistical advice for their research projects and theses and offered hands-on training in applying advanced research methods.

Cohen was a gifted award-winning teacher, as evidenced by the multiple certificates covering her Technion office wall. With her engaging style, she was able to clearly explain complex topics. Her video lecture series quickly became a classic among generations of Technion students and was later made available on YouTube. Throughout the years, she delivered many workshops on various topics, including structural equation modeling, mixed models, and moderators and mediators.

Cohen was an extraordinary mentor who inspired her students and collaborators to view and use statistics as a tool to improve people’s quality of life and change the world for the better. She was a true leader in making statistics accessible, useful, and fun. Cohen was a big advocate of learning from experience and always emphasized that mistakes are inevitable and useful in the learning process. Her love of research and excitement about the use of statistics to solve real-world problems will continue to inspire those who were fortunate to learn from and collaborate with her. She possessed a unique ability to create and maintain lifetime and ongoing relationships with many of her students in Israel and around the world. She was instrumental in making colleagues and students new to the faculty feel welcome and enthusiastic about statistics.

After her retirement in 2008, Cohen continued to teach, direct the statistics lab, and conduct research. She secured research grants, including one from the prestigious US-Israel Binational Science Foundation. A few days before her passing, she submitted her last research proposal.

To read more about Cohen, visit the blog created in her honor.

Brian (Brett) O’Hara

Brian J. (Brett) O’Hara, 47, a resident of Crofton, Maryland, died September 16, 2018, of a malignant brain tumor.

Brett was born in Libertyville, Illinois, and pursued his undergraduate studies at St. Olaf College. He was pursuing his doctorate in economics at the University of Notre Dame in 1995 when the oligodendroglioma was discovered. Brett underwent surgery, chemotherapy, and radiation and returned to his studies, earning his PhD in 2000.

He was passionate about health and disability research, specifically the economic effects of major illnesses and health insurance coverage. Brett worked for the federal government for 19 years, 16 of those at the US Census Bureau.

Brett was a loyal and true friend to many; however, Brett’s greatest joy was being a father to his children, Eva (18) and Aengus (12).

In lieu of flowers, memorial contributions may be sent to the Mandrin Inpatient Care Center, Hospice of the Chesapeake.

Edward Hugh Simpson

Chris Christensen, Northern Kentucky University

Edward Simpson—codebreaker, statistician, and civil servant—died February 5, 2019, at the age of 96.

Edward was born on December 10, 1922, and grew up in Northern Ireland as an only child. He attended Colerain grammar school and planned to study languages at university. However, in the summer of 1938, the headmaster suggested to Edward that he leave school a year early and go to Queens University, Belfast, to study mathematics because war was likely and the country would need mathematicians. So, at age 16, Edward went to Queens. He graduated with a 1st in 1942 at age 19.

In the fall of 1942, Edward was recruited to work as a codebreaker in the Italian Naval Section of Bletchley Park. In the fall of 1943, as the war with Italy was ending, Bletchley Park’s Italian Naval Section found themselves with little work and most of the codebreakers were transferred to work on Japanese ciphers. Edward was selected to lead Bletchley Park’s attack on JN-25, which was the primary Japanese naval cipher.

When the war ended, Edward went to Christ’s College Cambridge to continue his study of mathematics. His tutor was Maurice Bartlett. However, in 1947, Edward left to enter the civil service. He served first in the treasury and then as principal private secretary to Lord Hailsham, Lord President of the Council, and Lord Privy Seal. Most of Edward’s post-war career was in the department of education and science, from which he retired in 1982 as deputy secretary. He received a CB (Companion of the Bath) in 1976 for his services to education.

In 1946, while still at Cambridge, Edward wrote “The Interpretation of Interaction in Contingency Tables” with Bartlett. It was published in 1952 in the Journal of the Royal Statistical Society at Bartlett’s request so he could refer to it. It was this paper that introduced Simpson’s Paradox, which demonstrates that a trend appearing when data is segmented into groups can disappear or reverse in the aggregate.

In 1949, he published “Measurement of Diversity” in Nature. That paper introduces the Simpson Index (of Diversity). Although Edward could not mention it at the time of publication, the index was based on the repeat rate (RR) in use in the Italian Naval Section at Bletchley Park (BP) when he arrived. In a March 26, 2014, email, Edward discussed how the idea for the index arose:

[The idea] is so simple that it will spring up spontaneously all over the place. Anyone who looks at an enciphered text will start by doing a frequency count of its letters or figures or whatever it comprises and will at once see that the sum of the squares of the frequencies measures the repeat rate, the probability that two items selected at random will be the same.

So, when I did my four terms of mathematical statistics at Cambridge … and chose to elaborate the RR, this did stem from BP. But not only that. I had read Udny Yule’s “Statistical Study of Literary Vocabulary” at BP when it came out in 1944 and been very excited by it. This made it easy for me to explain my interest in the problem without mentioning BP. It is odd that my name got attached to the index (which I didn’t know for 50 years) when the name of Yule’s Characteristic was ready to hand.

What prompted me to publish it in 1949? I think just that I was two years into my civil service career, enjoying it, not going back professionally to statistics, and that it would be a pity to let this bit of work go to waste.

In 1947, Edward married Rebecca Gibson, another Bletchley Park codebreaker. Rebecca died in 2012. Edward is survived by a son, a daughter, and four grandchildren.

Babu Shah

Complied by Akhil K. Vaish, RTI International, from various sources and in consultation with J. N. K. Rao and Kanti Mardia


    Babu Shah, former chief scientist at the Research Triangle Institute (RTI International) and father of RTI’s SUDAAN software systems, passed away January 27, at the age of 83.

    Described by a former colleague as “the petite fellow from India,” Babu was one of RTI’s statistical giants. He retired from RTI in 2003, after 37 years of service to the institute. He held a bachelor’s in mathematics (1955), a master’s in statistics (1957), and a doctorate in statistics (1960) from the University of Bombay, India. His PhD thesis led to five brilliant single-authored papers on construction and analysis of experimental designs in the Annals of Mathematical Statistics (September 1958–June 1960). Oscar Kempthorne was so impressed by the elegance and beauty of Babu’s published research that he invited Babu to work with him as a research associate at Iowa State University (1959–1962). He worked on optimization in response surface methodology and jointly developed the ingenious method of parallel tangents (PARTAN), which is scale invariant and much faster than the well-known steepest descent method.

    Babu was a recognized international expert in statistics, methodological research, and data management. In addition to serving as chief scientist, he held positions at RTI as associate director, department manager, and statistician. He was an American Statistical Association Fellow, an elected member of the International Statistical Institute, and a Royal Statistical Society Fellow. He served as an adjunct professor in the department of biostatistics at The University of North Carolina at Chapel Hill.

    In 1984, RTI introduced Babu’s SUDAAN, the first statistical software created to analyze correlated data collected from complex surveys. The tool quickly became the industry standard, and the 11th version was released in 2012. Babu named it SUDAAN for two reasons—a shorthand acronym for “survey data analysis” and because the word, in Sanskrit, means “beautiful gift.” RTI’s small-area estimation software development would not have been possible without Babu’s inspired contributions.

    After his retirement, Babu started working on SAFAL (Statistical Analysis Functions Application Language). He was motivated to help graduate students and research statisticians who have to spend considerable time writing computer programs to evaluate their research formulae. SAFAL is a multiprocessor language that directly translates mathematical formulae into computer programs. Up until his death, with limited eyesight and 1-inch font, he continued this work and recently completed a working SAFAL prototype.

    On numerous occasions, Babu donated his paid time off to staff in need and provided financial assistance to his relatives and friends. He was unselfish in providing technical and statistical guidance to the hundreds of statisticians who have worked at RTI during his tenure. He was also one of the first to receive an RTI Mentoring Champion Award in 1999. Babu is fondly remembered by his former colleagues and friends:

    “I consider Babu’s contributions to statistical theory and methods, computation methods, and software development to have been absolutely critical and invaluable to the advancement of the field of survey research. In addition, his work and encouragement were very important to my career. He was a wonderful colleague and friend. I will cherish the memories of his unending energy and creativity when we worked together during the 1980s and 90s on the development of SUDAAN and of the statistical methods it implemented.”
    ~Barry I. Graubard, National Cancer Institute

    “Babu was one year junior to me in Bombay University. I graduated in 1956 and Babu in 1957. After spending two years as a research scholar in India, I came to Iowa State to do my PhD. Babu took the research scholarship in Bombay and, within two years, wrote a brilliant PhD thesis on balancing designs and related topics. I believe Babu is among the most original researchers and always came up with brilliant ideas. His work in experimental designs for his PhD thesis is path breaking. Babu’s contribution to SUDAAN is monumental. I was even looking forward to his breakthroughs in computing using AI ideas.”
    ~J.N.K. Rao, Carleton University

    “Babu was my close classmate at the University of Bombay (1955–1957). The number of students in the class were restricted to 24 and were selected from all over India. He was a quick thinker with great critical ability. He had a calm, balanced, and peaceful personality. He was a quiet and unassuming friend, always to the point, and, whatever was happening, he would take in his stride, which I very much admired. He came to Leeds University as a consultant to the experimental design and analysis package SELINA, which the department was developing in the 1980s. His great expertise led to a new direction to make the design and analysis of experiments more user-friendly and accessible to nonstatisticians. This software is still active, which is amazing over 25 years later.”
    ~Kanti V. Mardia, University of Leeds

    “The importance and utility of Babu’s statistical innovations and advances directed to the analysis of complex survey data is readily apparent in the vast number of citations attesting to his multifold contributions to the field. I was quite fortunate to work with and benefit [from] Babu’s expertise over several decades. His continuous commitment to the implementation of new advances in application via SUDAAN has facilitated a multitude of innovative model-based analyses of national health and health care data that were and remain essential to informing health policy and practice.”
    ~Steven B. Cohen, RTI International

    “Babu was creative, thoughtful, and unselfish. He had remarkable integrity; his smile was infectious (and he smiled frequently—thank goodness!). He was a wonderful mentor, a statistician’s statistician, and the epitome of a gentleman and a gentle man.”
    ~Kerrie Boyle, RTI International (retired)

    “Babu Shah has been my teacher, colleague, and friend for over 25 years, since I first joined the SUDAAN project. From him, I have learned how to take credit for my mistakes as well as my accomplishments and the importance of family above all else. Of course, there were the technical insights, the ingenious ideas for solving difficult problems, the staggering wealth of knowledge, and the unwavering certainty that no problem was too tough to crack that he regularly shared with me. But there was so much more to Babu. He was insatiably curious, unfailingly kind, and always generous with his time and ideas. No problem or idea was ever too small or too large to discuss with Babu, and despite his eminence and stature in the statistical community, he always treated my questions and ideas with respect. The thing that got to me more than anything else though was Babu’s intellectual honesty and unvarnished willingness not just to own his mistakes, but to celebrate them. I can’t count the times he sought me out and gleefully reported a “booboo” he had made. Babu once said that the word SUDAAN means “beautiful gift.” I know in my heart that Babu was our beautiful gift.”
    ~Beth Barnwell, RTI International (retired)

    He was the baby of the family. With two older brothers and four older sisters, he was the youngest of the youngest. He wasn’t supposed to be the patriarch of the family, but with the premature passing of his older brothers, he assumed the role with grace, humility, and brilliance.

    What a polymath he was. No DIY project was too daunting to undertake, no circuit diagram was too complex to decipher, no puzzle was too hard to solve, no origami object was too hard to master. He was indeed our beautiful gift.

    Babu is survived by his wife, Ketki Shah; son, Parag Shah; and daughter, Mona Shah-Shurland.

    1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)

    Comments are closed.