George Box: A remembrance

Professor Box during an evening of pizza and beer at JMP Discovery Summit in 2009

Today I mourn the passing of George Box, truly a giant among 20th century statisticians. Indeed, I should not relegate him to the 20th century because he won the Brumbaugh Award in 2010 and 2007 for writing papers that made in their year of publication the largest single contribution to the development of industrial application of quality control.

Professor Box was 93. The last time I saw him was at the JMP Discovery Summit conference in 2009 where I introduced him to give a speech. George got a standing ovation from a crowd of several hundred fans of design of experiments and particularly his work. I will never forget his remarks as the applause died slowly away.

He said, "I feel like the son of the sultan on his 21st birthday when presented with 21 virgins. I know what to do. I just don't know where to start!"

Professor Box was the father of response surface methodology, which arose following his groundbreaking 1951 paper with K.B. Wilson. He also popularized the regular two-level fractional factorial designs, writing two great papers with Stu Hunter in 1961.

His greatest contribution to my life was the wonderful book, Statistics for Experimenters, which he wrote with William G. Hunter and Stu Hunter and published in 1978, the same year he served as president of the American Statistical Association. I remember the excitement I felt on reading the description of how the attainment of knowledge is an endless spiral proceeding alternately from deduction to induction and back. Even now, I recall with pleasure the discussion of the randomization distribution early in the book. Of course, the chapters on the two-level fractional factorial designs gave me the tools that resulted in my first big successes as an industrial statistician.

Because of my passion for design of experiments, I tend to focus there. But I tip my hat to his contributions in Bayesian analysis (Box and Tiao 1973), time series (Box and Jenkins 1970) and control (Box and Luceño 1997).

Before I introduced him at the conference in 2009, I asked him what his favorite honors were. He mentioned that he had become a Fellow of the Royal Society. He made sure I knew that was different (and much better) than being a Fellow of the Royal Statistical Society, where he was made an honorary Fellow in 2004. He said that to become a Fellow of the Royal Society was more prestigious than being knighted.

I first met him at a Gordon Research conference in the late 1980s. As a young statistician, I am afraid that I was a bit too overawed to take full advantage of that opportunity for knowing him on a personal level. But listening to his insights in the discussion after each talk at the conference was, by itself, worth the week spent sleeping in the terrible beds at a boys' school in New Hampshire.

Yes, the world of statistics has lost a great one. I will miss him. Requiescat in pace, my friend.

Box and Wilson K. B. (1951) "On the experimental attainment of optimum conditions." J. Roy. Statist. Soc. Ser. B Metho. 13:1-45.

Box, G. E. P. and Hunter, J. S. (1961) "The 2k-p Fractional Factorial Designs Part I" Technometrics Vol 3, No. 3 311-351.

Box, G. E. P. and Hunter, J. S. (1961) "The 2k-p Fractional Factorial Designs Part II" Technometrics Vol. 3, No. 4, pp. 449-458.

Box, G. E. P. and Jenkins, G. (1970) Times series analysis. Forecasting and control. Wiley, New York.

Box, G. E. P. and Tiao, G. (1973) Bayesian inference in statistical analysis Wiley, New York.

Box, G. E. P., Hunter, W.G., Hunter, J. S., (1978) Statistics for Experimenters, In Introduction to Design Data Analysis and Model Building. Wiley, New York

Box, G. E. P. and Luceño, A. (1997) Statistical control by monitoring and feedback adjustment Wiley, New York, 328 pp.

tags: Design of Experiments (DOE), Discovery Summit, Innovators' Summit, Statistics

19 Comments

  1. Posted March 29, 2013 at 9:54 am | Permalink

    Very well written, Brad. Thanks for this. I only met Dr. Box once - at the JMP Discovery conference in 2009 that you had referenced. Like your first encounter, Brad, I was rather awestruck and terribly grateful. I asked him how gratifying it must be to have made the enormous contribution he has over his career. I told him he had a profound effect on me. His work gave me purpose by giving me tools to do important things for others. "It would be impossible to calculate the enormous gains around the world that are rooted in your methods, Dr. Box" I said to him. For a few moments, he was completely silently, looking directly at me, his eyes welled-up. He reached out and gave me quite the hug. It was quite a moment. Such a humanitarian. He deserved what he got - a long life of enormous accomplishment. The world is a much better place because of Dr. Box.

    • Paul Hollingworth
      Posted April 4, 2013 at 9:03 am | Permalink

      Wayne, thank you for sharing that. I never met George Box in person, a matter of personal regret. I feel connected to him now through you my dear friend. An extraordinary life.

  2. Posted March 29, 2013 at 11:32 pm | Permalink

    Brad,

    Nicely said!

    I grew up in Madison, Wisconsin as the son of William G. Hunter (who followed George Box there to become one of his first PhD students and later a professor there at the Statistics Department that Box founded and co-author of Statistics for Experimenters). Growing up as a kid, I had no idea how influential George Box was in his chosen field(s). All I knew was he was a really fun friend of my parents to be around because he was an absolutely fantastic story teller with a great sense of humor who was always quick with a smile or a quip and always seemed to be in a happy mood. People just naturally gravitated to him at parties because of his charm and enthusiastic willingness to chat. The twinkle we would frequently get in his eye when he told stories was irresistible.

    When I got older, my appreciation for him grew. After my dad died in 1986 and a statistics conference was dedicated to him a year or two later, George Box gave a speech describing my dad's contributions in the field and his qualities as a person that I consider, to this day to be the nicest, most thoughtful, most insightful comments I've ever heard about anyone make about someone I've known well.

    Later still, after I had gone to college and had developed an appreciation for how excellent professors with a real gift for explaining things well were dramatically more enjoyable to listen to and easy to learn from as compared to simply "very good" professors, I had the pleasure of listening to George Box talk about advances in statistics throughout the years. It was remarkable. Not only was he funny and insightful, I was hugely impressed by his ability to construct his stories. I've never seen anyone do what he did that night as well as he did. As a listener, I found myself going from "OK, really interesting background details to set up the problem, but... where is he going with this?" to "Wow! That's such a clever and elegant solution." What made such a big impression on me was the journey he took listeners on as he progressed towards the solution, what he said, the things he left unsaid, his humor, and his timing were all wonderful. My dad had often said a big reason that he loved hearing Box teach was that he truly had a gift at making complex ideas unfold clearly. Listening to Box that night, I understood what he had meant. It was like watching Michael Jordan at his prime; a virtuoso performance that left people in attendance feeling grateful to have been able to witness something special.

    Since hearing about his death, I downloaded his autobiography and read it until 3:00 AM last night. So far, it's everything I would expect. It's called "An Accidental Statistician: The Life and Memories of George E. P. Box" http://www.amazon.com/An-Accidental-Statistician-Memories-ebook/dp/B00BU8Z3R6 It includes very funny stories told by a master story teller.

    Justin Hunter

    • Bradley Jones Bradley Jones
      Posted March 30, 2013 at 7:45 am | Permalink

      Justin,
      Thanks for your tribute and for the link to the autobiography. I just downloaded it from Amazon. I am so glad that Dr. Box finished it.
      -Brad

      • Posted April 8, 2013 at 11:11 am | Permalink

        Dear Brad and Justin, your tributes to Prof Box gave people like me an insight into the life of this great statistician. It is from your articles that I learnt he was also a great human being. I am an amateur statistician who got to (or rather was forced to) learn a lot of statistical methods during my tenure in GE; the name Box used to keep cropping up in most of the stuff that we had to read, and I started to wonder what kind of a person he would be. Reading your pieces gave me a complete picture. Like truly great scientists, his contributions to his chosen field positively influenced many people in unrelated professions.

  3. Posted March 30, 2013 at 12:03 am | Permalink

    Brad,

    Nicely said!

    I grew up in Madison, Wisconsin as the son of William G. Hunter. My dad followed George Box there to become one of Box's first PhD students. Later, he became a professor there at the Statistics Department that Box founded and he co-author of Statistics for Experimenters with Box. Growing up as a kid, I had no idea how influential George Box was in his chosen field(s). All I knew was he was a really fun friend of my parents to be around because he was an absolutely fantastic story teller with a great sense of humor who was always quick with a smile or a quip and he always seemed to be in a happy mood. People just naturally gravitated to him at parties because of his charm and enthusiastic willingness to chat. The twinkle he would get in his eye when he told stories was irresistible.

    When I got older, my appreciation for him grew. After my dad died in 1986 and a statistics conference was dedicated to him a year or two later, George Box gave a speech describing my dad's contributions in the field and his qualities as a person that I consider, to this day to be the nicest, most thoughtful, most insightful comments I've ever heard about anyone make about someone I've known well.

    Later still, after I had gone to college and had developed an appreciation for how excellent professors with a real gift for explaining things well were dramatically more enjoyable to listen to and easy to learn from as compared to simply "very good" professors, I had the pleasure of listening to George Box talk about advances in statistics throughout the years. It was remarkable. Not only was he funny and insightful, I was hugely impressed by his ability to construct his stories. I've never seen anyone construct their stories as well as Box did in that presentation. As a listener, I found myself going from "OK, really interesting background details to set up the problem you're discussing, but... where are you going with this?" to "Wow! That's such a clever and elegant solution." What made such a big impression on me was the journey he took listeners on as he progressed towards the solution (what he said, even more importantly - the things he left unsaid, his humor, and his timing) were all wonderful. My dad had often said a big reason that he loved hearing Box teach was that he truly had a gift at making complex ideas unfold clearly. Listening to Box that night, I understood what he had meant. It was like watching Michael Jordan at his prime; a virtuoso performance that left people in attendance feeling grateful to have been able to witness something special.

    Since hearing about his death, I downloaded Box's very-recently-published autobiography and read it until 3:00 AM last night. So far, it's everything I would expect. It's called "An Accidental Statistician: The Life and Memories of George E. P. Box" http://www.amazon.com/An-Accidental-Statistician-Memories-ebook/dp/B00BU8Z3R6 It includes very funny stories told by a master story teller.

    Justin Hunter

  4. Posted March 30, 2013 at 3:59 am | Permalink

    Brad's testimonial is combining the methodological, educational, practical and personal contributions of this giant whose impact on statistics and other disciplines cannot be understated - thanks Brad.

    Two minor add-ons are worth mentioning. A concretisation of the above were the beer and statistics evening seminars at his home in shorewood hills in madison. these unique events combined theory and practice with an emphasis on practical statistical efficiency, lots of good humor and of course beer - an ideal environment for learning and innovation.

    My personal Box anecdote has to do with social security numbers. The one i have is particularly strange because it combines a set of repeated descending numbers. When i mentioned that at one of the lunches in Madison, either Jerry Kloz or George Tiao said: based on my SS number, i always thought that the first three numbers corresponded to the area code. George (Box) then immediately said: you cannot make inference on the basis of a sample of size 1...

    George was the ultimate statisticians combining contributions in mathematical developments with statistical thinking at unprecedented levels. As a profession we lost one of our founding fathers. His impact as role model will surely continue to be felt. My deepest condolences to Claire.

    • Bradley Jones Bradley Jones
      Posted March 30, 2013 at 7:53 am | Permalink

      Ron,
      Thanks for contributing your tribute here. I hope that other statisticians will also chime in.

      Of course people make inferences on the basis of a sample of size one all the time. :-)
      Not very good ones in most cases I suspect.
      -Brad

  5. Posted March 31, 2013 at 5:35 am | Permalink

    I never met Prof Box, but he was one of the few statisticians whose work I was familiar with as a biomedical scientist, and whose tools actually made a difference to my analysis of REAL datasets. So much of statistics is concerned with idealised data and assumptions that are commonly violated by real-world data (at least biological data!) but statisticians like George Box and Svante Wold gave me hope that we could extract meaning from large, complex datasets. Even today, we are only part way along that journey, but it was a journey that only started because of a handful of great men with the perspicacity to see what was required - and George was among the vanguard.

  6. Posted April 1, 2013 at 1:12 pm | Permalink

    Brad,

    Very nice article and tribute. I did not know Mr. Box and have only talked to John Hunter and Brian Joiner about him. Those conversations made a lasting memory on me, as the respect and admiration they held for Mr. Box was touching. A great teacher and a pioneer in the world of statistics.

  7. Jim Duarte
    Posted April 1, 2013 at 3:46 pm | Permalink

    All of the tributes that have been made are wonderful. I had the opportunity over the years to meet George at many ASA/JSM meetings. The most impressive meeting was when he came to HP to teach a class on frational factorial designs (in the days of the Taguchi popularity) with Conrad Fung and Soren Bisgard. The trio were very good educators and entertaining. During my manufacturing career I used a small book by Box and Draper quite bit: 'Evolutionary Opertions'. It was like DOE "on the fly". Combining that with Response Surface analysis made a tool set that went a long way in improving the processing of aluminum plate, sheet and extrusions then later at HP in printed circuit board manufacturing. George was a very enjoyable person as was Bill Hunter. Bill also came to HP for a few day and was so personable and helpful to us statisticians. Justin, you should be very proud of your father's contributions also. We've lost some great ones, but their contributions will live long after them.

  8. Posted April 3, 2013 at 1:37 am | Permalink

    I got to know George Box well, first as a young Ph D student working under his guidance and later as a frequent visitor to Madison.

    Doing Ph D research with George was a remarkable learning experience. He gave us a great deal of intellectual freedom to pursue our own ideas; and in parallel engaged us in his deeply held view that statistics, and statistical research, must derive its raison d'etre from real problems encountered in science and engineering. He was fond of mid-day walks to picnic point (a scenic spot on the UW campus) and some of our best discussions of my research came on those walks.

    Working with a brilliant thesis supervisor is always somewhat intimidating. We always had a sense that, whatever we did, George "already knew it". I remember one week when I had figured out some really neat connections between some Bayesian models and my office mate had discovered some surprising results analyzing a data set. We were excited about showing off our discoveries to George; and he was pleased with our work. And although he never said so directly, we both came away knowing that he had already been there.

    One of the great traditions at Wisconsin was George's Monday night beer and statistics symposium. These meetings, held in the basement of his house, were devoted to discussing applied problems presented by guest researchers. George supplied the beer and cider; as his students, we brought in the speakers and one of our most important tasks was to stop at the liquor store to pick up the beer. For many students, the Monday night meetings provided their first encounters between science and statistics and it was a wonderful way to see statistics in action. One component of that learning was to see a master in action. George was superb at cutting to the heart of problems and leading the researcher to useful suggestions and ideas. He had incredible instinct for crystallizing the key questions and focusing the directions to answering them -- an instinct that became especially sharp once he had downed a couple beers!

    On a personal level, George was warm, friendly and always with great stories to tell. He had fantastic stage sense -- statistics gain was the theater's loss.

    With George's death, I and his fellow students have lost both a mentor and a dear friend.

    • Bradley Jones Bradley Jones
      Posted April 3, 2013 at 6:03 am | Permalink

      David - now that you have your own students, you are the one who has to be careful about the intimidation factor. I suspect that by now you have already been there many times. Thanks for your tribute. -Brad

  9. Bovas Abraham
    Posted April 3, 2013 at 9:47 pm | Permalink

    George Box was a giant in Statistics, also a mentor and friend to many students including this one.
    My first meeting with George was in September 1971 when I joined the University of Wisconsin-Madison as a new graduate student in Statistics. I was assigned as a research assistant (RA) for Professor George Box. I had only heard of the name George E.P. Box and never met the man. Some fellow students advised me that there was no need to see him because he had several RA’s and he might not bother with a new arrival. However, I made an appointment to discuss the RA duties. In that meeting he asked me whether I could draw some “HPD regions”. I had no idea what an HPD region was but I was not very anxious to reveal my ignorance. In any case he gave me some details and I left the office thinking about the new adventure. That was the time of punched cards, calcomp plotters etc. After consultation with other statistics graduate students and some people from computer science I found a program which was adapted to draw the required HPD region and George Box was very happy. That was the beginning of a long friendship which I treasure. I was fortunate to be his Ph. D student and also to assist in the books “Box-Hunter-Hunter” and the revised edition of “Box Jenkins”. This association changed the way I thought about Statistics. I always wonder about my good fortune to be associated with him.
    One of the things that keeps coming back to my mind is the Monday night "Beer Sessions" held in George's basement where research problems were discussed on an onging basis. It always started with somebody talking about a data set.

    George had a keen sense of humor. He used to dress up at departmental christmas parties, participate in skits, and make up humorous songs about statistics, Bayesians etc
    "There is no theorem like Bayes' theorem
    Like no theorem we know...."

    Once George and about seven of us were leaving from a conference in Tel Aviv to the ISI conference in Cairo. We had to report at the airport 4-5 hrs early for the flight and it was an early morning flight. The immigration procedures at the airport were extremly slow and after 2 hrs we moved about 20 ft and we had a long way to go. George's comment - "If there ever was a stationary process, this has to be it."

    Many of us are familiar with his famous statement "All models are wrong, some are useful".

    Here is another one, "The quality movement can be seen as the the analyis, institutionalization and democratization of Scientific Method, a tool for efficiently generating new knowledge." There are many such gems.

    George E.P. Box was one of the most respected statisticians, and a legendary figure among his peers. However, perhaps his greatest contribution to the discipline has been his perspective of Statistics as a philosophy --- the tangible essence of an iterative learning process. To George Box, statistical tools and methods provide the logic and the language of the modern scientific method. George Box has greatly influenced consecutive generations of researchers and practitioners, not only in his own field of statistics but also in engineering and industry. So his legacy will continue through his students and associates.

    Thank you, Thank you George. Bye my friend.

  10. TN Goh
    Posted April 4, 2013 at 9:32 am | Permalink

    Let me just add this. Lucky are those who had George as advisor. But how many know that George would spend hours, one-on-one, enlightening a shy research student from outside the Statistics Department? (That student was me). And yes, those Monday night beer sessions! ... I left Madison, WI forty years ago, but till today I still remember vividly his sharp remarks and jovial songs, whenever called for by the occasion!

    • Yao Yiyang
      Posted April 8, 2013 at 3:04 am | Permalink

      What a precious memory~ I also hope I would have such a tutor in my life.

  11. Manny Uy
    Posted April 4, 2013 at 9:40 pm | Permalink

    Thank you all for posting your experiences with a true giant in scientific thinking. Like Brad and Wayne, I knew of GEP Box from the Box, Hunter and Hunter textbook. I only saw and heard him once at the JMP Discovery Summit in Chicago. I envy those who had participated in the many beer sessions with him, and could only imagine the fun after two or three glasses. The closest I have ever imagined about those beer sessions was a similar beer (and wine) evening with Stu Hunter at the hotel pub during an earlier JMP Discovery Summit in Cary. That was when I learned that Box was also the son-in-law of Sir Ronald Fisher! Thank you all for the other stories revealed in this testimonial. I will continue to learn more about him by reading the "Accidental Statistician". Thank you Brad for pointing out that his contributions extended into the 21st century, and for the list of significant publications that summarized his professional life but not the personal one like those written above.

  12. Barbara Nelson
    Posted April 26, 2013 at 8:53 pm | Permalink

    Nicely said Brad. George Box died on my Dad, Lloyd Nelson's 91st birthday. Unfortunately he is in the late stages of Alzheimer's disease. I'm sure that Dad also met George Box at the Gordon Research conference and had many a lively discussion.

    • Bradley Jones Bradley Jones
      Posted April 27, 2013 at 6:52 am | Permalink

      Barbara - thanks for your note. I am grateful to your father for being the founding editor of the Journal of Quality Technology. I hope that I can live up to his vision in my stint as editor of that wonderful journal.

3 Trackbacks

  1. By Predictive modelling returns to the UK on April 22, 2013 at 9:49 am

    [...] predictive modelling methods can be used to make the models built by JMP Pro as useful as possible. George Box is one of the great statisticians of the last century and sadly passed away earlier this month, [...]

  2. By Celebrating statisticians: George E.P. Box on May 6, 2013 at 10:18 am

    [...] George passed away in late March, and I wrote a remembrance of him for the JMP Blog at that time. That blog post expresses what I would have written in a post [...]

  3. […] Bradley Jones posted his thoughts on George – remembrance by Professor Bovas Abraham – En Memoria de George E. P. Box, Doctor […]

Post a Comment

Your email is never published nor shared. Required fields are marked *

*
*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <p> <pre lang="" line="" escaped=""> <q cite=""> <strike> <strong>