Kjell Doksum

Submitted by Vijay Nair

Kjell Doksum, an ASA Fellow, died November 20, 2021, at the age of 81. He was professor emeritus of statistics at the University of California, Berkeley, and senior scientist in the statistics department at the University of Wisconsin, Madison.

Kjell was born in Sandefjord, Norway, and grew up in Oslo. His parents were Filip Doksum and Elise Olsen, and he had two brothers. He was married to Joan Fujimura, a distinguished scholar in the sociology of science, and had three daughters: Teresa; Kathryn; and Margrete.

Kjell’s lifelong passion was soccer (or football). He grew up near Bislett Stadium and started playing in the streets of Oslo. Alas, his skills on the cobblestones did not translate to the pitch and his hopes of becoming a professional player were dashed. Nevertheless, he remained an avid soccer fan and played recreational soccer well into his ’60s. Generations of students, faculty, and visitors at Berkeley, Madison, and elsewhere will fondly recall the practices, games, and parties he organized. It was during one of the soccer parties that I asked him if I could work with him on my dissertation. Not being a man of many words, Kjell just said, “Sure.” He became not only my adviser, but lifelong mentor and close friend.

Another potential career that didn’t work out was fishing. After high school, Kjell moved to San Diego to join his aunt and uncle. Many of his relatives were fisherman, so he tried it out for two weeks. After getting seasick from the huge waves during a storm, he decided that going to San Diego State College was a smarter and safer option. The fishing industry’s loss was a huge gain for statistics.

Kjell earned his master’s degree in statistics from San Diego State College in 1962, where he worked with Chuck Bell, one of the few African American statisticians at the time. He then moved to the University of California, Berkeley, where he completed his PhD in 1965 under the supervision of Erich Lehmann. After a year as a postdoc in Paris, he joined the Berkeley statistics department, where he spent most of his academic career. He took an early retirement in 2002 and moved to the University of Wisconsin, Madison, with his wife, Joan, who was recruited by the department of sociology. Kjell retired from Wisconsin in 2010 but remained there as a senior scientist. He held visiting positions at the L’Universite de Paris, University of Oslo, the Norwegian Institute of Technology in Trondheim, Harvard University, Columbia University, Bank of Japan, Hitotsubashi University in Tokyo, and Stanford.

Kjell made pioneering contributions to statistical theory and methodology, covering a wide range of areas: randomization methods; nonparametric and rank-based inference; survival and reliability analysis; semiparametric techniques; transformation models; probability measures; and Bayesian inference. In early collaboration with Bell, he developed a randomization procedure and showed the resulting Gaussian randomized tests were asymptotically more powerful in non-normal situations than the classical ones. He returned to this topic later, when he proposed a Monte Carlo approach for partial likelihood inference for semiparametric models.

In the early 1970s, Kjell introduced the shift function as a nonparametric functional measure of difference between distribution functions and, in a series of subsequent papers, developed inference procedures. In fact, he was still working with a student on extending this approach to regression settings at the time of his death.

In one of his most cited papers, Kjell proposed the well-known “neutral-to-the-right” processes and developed their properties. These processes are useful in nonparametric Bayesian inference. Kjell’s joint work with Peter Bickel provided deep insights into statistical properties of procedures based on transformed data. In joint work with Steinar Bjerve and others, he introduced the concept of local correlation in nonparametric regression framework. He proposed, jointly with Arnljot Hoyland, a new class of degradation models for reliability and life testing. He and his coauthors also made many important contributions to survival and censored data analysis. The Festschrift volume organized for his 65th birthday, Advances in Statistical Modeling and Inference: Essays in Honor of Kjell A. Doksum (2006), edited by V. Nair, covered contemporary results on all these research areas.

Many graduate students at Berkeley and elsewhere learned their statistics theory and methods from Kjell’s book, coauthored with Peter Bickel, Mathematical Statistics: Basic Concepts and Selected Topics. I fondly remember the early edition (called the ‘Red Book’ by students), all the more because of the many errors we had to work through. It made our learning more robust!

Kjell supervised 24 PhD students and several undergraduates. Quite a few returned to their home countries to pursue successful careers. He served on the editorial boards of JASA, Scandinavian Journal of Statistics, Life Data Analysis, and Sankhya. He was executive secretary of the Institute of Mathematical Statistics (IMS) and played a key role in the founding of the IMS journal Statistical Science. He was recognized for his contributions by being elected a fellow of ASA and IMS, an elected member of the International Statistical Institute, and a foreign member of the Royal Norwegian Society of Sciences and Letters.

Kjell was a kind, gentle, and soft-spoken person. He always had a sweet smile on his face, a great sense of humor, and time for everyone. He rarely talked about himself, except to Joan who used her ethnographic interviewing skills to get him to talk about his childhood and views on all things political. Kjell was a fierce advocate for social justice. He had developed an integrity and moral strength that carried him through many adversities, beginning with his mother’s early death and the Nazi occupation of Oslo during his early childhood.

Leland Wilkinson

Leland Wilkinson, H2O.ai chief scientist and longtime ASA member, passed away December 10, 2021, following a stroke.

During the late 1970s and early 1980s, Leland wrote SYSTAT, the first comprehensive statistical software package designed expressly for microcomputers. It represented an end-run around the punch cards, queues, and mainframes required for statistical analysis at that time. The program was the first of its kind to include comprehensive graphics driven by a command structure of universally applicable options, foreshadowing the graphical structure Leland would more fully develop and articulate during the 1990s. SYSTAT was also the first software implementation of the now widely used heatmap display.

Leland founded SYSTAT, a company of the same name headquartered in Evanston, Illinois, and sold it to SPSS in 1995. He went on to build a team of graphics programmers there who developed the nViZn platform that produces the visualizations in SPSS, Clementine, and other analytics services.

Leland wrote the seminal book on statistical graphics, his magnum opus The Grammar of Graphics, in 1999. The Grammar of Graphics provided a new way of creating and describing data visualizations, a language—or grammar—for specifying visual elements on a plot, which was a completely novel idea that has fundamentally shaped modern data visualization. The book served as the foundation for the R package ggplot2, the Python Bokeh package, and the R package ggbio. It also helped shape the Polaris project at Stanford University.

Leland earned a bachelor’s degree from Harvard, a bachelor of sacred theology degree from Harvard Divinity School, and a doctorate from Yale. He was an adjunct professor of computer science at the University of Illinois at Chicago, a fellow of the American Statistical Association, an elected member of the International Statistical Institute, and a fellow of the American Association for the Advancement of Science. He won the best speaker award at the National Computer Graphics Association and the Youden prize for best expository paper in the statistics journal Technometrics. He also served on the Committee on Applied and Theoretical Statistics of the National Research Council, was vice chair of the National Institute of Statistical Sciences Board, and served on the Board of Trustees of the Institute for Pure and Applied Mathematics at The University of California, Los Angeles.

To learn more about Leland’s life and work, visit his Wikipedia page.