Stan Azen: A Master of Numbers and Notes

Computational Statistics and Data Analysis 131 (2019) 2–9 Contents lists available at ScienceDirect Computational Statistics and Data Analysis journal homepage: www.elsevier.com/locate/csda Interview Stan Azen: A Master of Numbers and Notes Stan Azen Interview April 23, 2017 [Revisions completed July 2018] Interviewed by Abdelmonem A. Afifi Transcribed by Jardee Transcription, Tucson Arizona Afifi: This is an interview of Dr. Stanley Paul Azen. Today is Sunday, April 23, 2017. My name is Abdelmonem A. Afifi. We are now starting the interview. So let’s start by asking you, Stan, to tell me about your childhood. Azen: So I was born in Indiana, but my mother’s family lived in California. My mother really wanted me to move from Indiana to California. Although at the age of 11 while living in Indiana, I was the youngest person admitted to the Jordan Conservatory of Music at the university in Indianapolis. That’s what opened me up to classical music. The teacher I had, who also ran the Conservatory, was a disciple of Rachmaninoff’s and so I learned classical music from him and that was really great. In those early days, people would go to one of the European towns and pay to meet and study with talented musicians like Rachmaninoff. This is what my teacher Ozan Marsh did, and it was really incredible. And then when I did come to California, many of the people I met were musicians, mostly Russian Jewish people who really knew that kind of classical music also. So I was trained really well in classical music, both in Indiana and in California. Afifi: So how old were you, Stan, when you came to California? Azen: I would have been 13 years old. Afifi: By then you were already accomplished in music? Azen: Yes. Music was my hot topic. And I learned Chopin and Brahms and Beethoven. That was the beauty of that period. And at one point I even had the honor of turning pages for Stravinsky himself, when he was conducting a concert in Los Angeles! Afifi: Fantastic! Shortly after, I imagine, you went to high school. Tell me about your high school days. Azen: I went to Fairfax High School in Los Angeles, and that was one of the reasons I moved from Indiana out to California. And my mother was my main backer; she wanted to make sure that I had a good education. In fact I’m in the ‘‘Hall of Fame’’ at Fairfax, for my contributions to music and science over the years. My mother also was very interested in education for herself. Although she faced some difficulties in earlier years while in Indiana, she succeeded by attending the University of California at Santa Cruz, beginning at the age of 68. She majored in Comparative Literature, obtaining her B.A., M.A. and then Ph.D. at the age of 78, in 1991! Education and literature were extremely important to her. In fact she published her autobiography, Moving Tales: My Journey from Victim to Victor [Shirley Primack Azen, 2001], at the age of 88, which detailed her early struggles and ultimate success. https://doi.org/10.1016/j.csda.2018.08.024 0167-9473/ Interview / Computational Statistics and Data Analysis 131 (2019) 2–9 3 Just about everybody at Fairfax High School at that time was from Europe, and there were a lot of children of Holocaust survivors there. So I heard French, I heard German, I heard Spanish, I heard all sorts of languages, and I found that to be really exciting. I started learning about other languages and cultures. And, Fairfax High School had a huge number of Jewish people, who then started taking me to their family homes for all the celebrations that Jewish people had. I was never given that by my mother. It was all secret, secret, secret — because the family, my grandparents, escaped from Russia during that Cossack period and the pogroms of 1903–1906. Afifi: So your mother’s parents had escaped from Russia, and she was hiding the fact that she was Jewish? Azen: That’s right. Afifi: But in Fairfax High School, all of that was wide open — was that something that established your identity as well? Azen: Yes. And that’s where I learned, really, what the whole Jewish world was about. And that’s where I was invited to appreciate breakfast or dinner in the Jewish way. And it did open me up to what it was to be Jewish. You see, in that little Indiana town, there were almost no Jewish people there. Afifi: Was it more the cultural aspects of it, or the religious aspects? Azen: I would say it was a mix of the two, both the cultural and religious. And I would probably put cultural first, and then the religious, because when I would spend time with people from that period, they would talk to me about the concepts of what it is to be Jewish. And then I did learn that my name, Azen A-Z-E-N, was a Jewish name given to my grandparents who learned that in Pittsburgh: there was a well-known furrier named Max Azen. If you go there and you go to the University of Pittsburgh, there’s a Tower of Learning and it’s dedicated to Max Azen. Afifi: Fascinating! Was Max Azen related to you? Azen: He wasn’t. I think they just took that name because there was a wealthy man named Max Azen. When my son Matthew got his degree in architecture, he met kids who had access to the Ellis Island documents. Through them he found out that our original family name was ‘‘Aisenberg’’, meaning ‘‘iron mountain’’ but as happened commonly in Ellis Island then, they changed it to ‘‘Azen’’. Afifi: So in high school, I guess you were thinking ahead of what career you wanted, and what you wanted to study in college. Tell me about that. Azen: Well at Fairfax, where I was in high school from 1952 to 1956, I met so many people who were creative in terms of understanding different kinds of strategies. They knew how to ask questions and so on, that opened me up to the idea that by going to college, I would hopefully be able to pursue research as a career. So I did go to UCLA [University of California, Los Angeles] in 1960, and that’s where I met you. Afifi: So as an undergraduate at UCLA, what excited you? I know you majored in math. Azen: Yes, mathematics was the one of the key topics. And actually, this is the kind of thing I was trying to figure out. The number was the thing. And then I would say probably also music in those early days — I wanted to be a master of both numbers and notes! Afifi: So you studied a mixture of mathematics and music? Azen: Yes. But then at the same time I started learning about research. You were a founding father for that at UCLA. You taught me about what it meant to do research. I remember you developed computer-based prognostic indices for critically ill patients. Your work even was the forerunner of what was later to be called logistic regression! And I also had the opportunity to work at the RAND Corporation in Santa Monica, California, which I did while I was a graduate student at UCLA. Afifi: O.K. So let’s talk about your graduate school days. I was witness to part of it, but tell me in your own words. 4 Interview / Computational Statistics and Data Analysis 131 (2019) 2–9 Stan with his Mom Shirley, graduating from UCLA with his Ph.D., 1969 Azen: Well, I know that your name always kept coming up as the important person. I was very thankful to be your second Ph.D. student in biostatistics — and now you’ve been chair or a member of more than 200 doctoral committees, that’s amazing! Afifi: Thank you. Azen: Graduate school would have been where I met a lot of the people who also were involved with music, and those who would explain to me how they do research. So they were very smart, but they would also say ‘‘You need to see Professor Afifi so he can explain that to you’’. If we talked about other early people who influenced me, Jerry Millstein would have been one of them. Jerry was the only other guy I met early on who knew music, and who also was into computer programming. When we first met he said to me, ‘‘Stan, you’re a great pianist, and I love your music, but I have to tell you, it’s pronounced] Bay-thoven, not Bee-thoven’’. I didn’t even know how to pronounce it correctly; I had been so isolated from others with musical interest! Another great early friend of mine is Ira Cooperman. He’s a great writer himself, and helped to inspire some of my earliest musical compositions. Afifi: That’s really something. Azen: When I worked at the RAND Corporation in the 1960’s, I learned about a kind of developing software called the ARPAnet, which was the precursor to what is now the Internet! So that whole Internet concept was developed at that same period, and at RAND I was one of the programmers of the ARPAnet. I had to obtain high level security approval to be able to work on the programs to do that kind of research, and make a sworn statement that I wasn’t going to misuse or reveal the code. So that was a really incredible project. I have pictures of me sitting on the beach by the RAND offices, playing a saxophone to relax. Afifi: So you did some work at Cal Tech [California Institute of Technology] and JPL [Jet Propulsion Laboratory] in Pasadena as well? Azen: Yes. And let me just tell you that one of the main things about working with JPL and Cal Tech at that time was that they were interested in what was being done in medicine, and they wanted to make contributions to advance medical research. For example, my involvement in cardiovascular research utilized a surrogate endpoint for vascular disease, by measuring the thickness of the carotid artery. I collaborated with faculty at JPL to utilize their imaging techniques designed for space to evaluate the human arteries! This imaging measurement was actually approved by the FDA [Federal Drug Administration] in the US as one of the first surrogate endpoints for cardiac disease, Interview / Computational Statistics and Data Analysis 131 (2019) 2–9 5 Afifi: You did your graduate dissertation on missing values in longitudinal data, as I recall. Azen: Right. And I like what you’re saying, because techniques like that I learned in graduate school, I see here in the book, Statistical Analysis: A Computer Oriented Approach [1979], in which some of the content topics include that of data measurements and computational tools, software, preparation of data for package programs, what to look for in a package statistical program, and other uses of the computer as a statistical tool. I see that I had dedicated this copy of it ‘‘To Mom, with love — my inspiration’’. And later the book was even translated into Russian, tying back into my family’s roots! Afifi: You’re talking about the book that you co-authored with me while you were still a student, actually. I remember that we would sit together and work on it over a glass of Jack Daniels whiskey! Azen: That’s right! For my talks, I actually put together a slide with a version of the book cover with you and me doing the work, and then Jack Daniels popping in as our third ‘‘co-author’’. Do you remember that? Afifi: I do remember that! And I remember the fascinating thing about that experience is that we would just sit and ramble, and you would take hand written notes, and then the next time we’d get together, you’d come back with organized notes that you had already actually typed and made into a smooth-reading chapter. I can never minimize your contribution to that pioneering book on how to use software for statistical analysis. Azen: Well, you know, you’re bringing tears to my eyes, because it brings back memories of our work together in the early days. And those were all incredibly helpful interactions — whereas I can picture in modern times, kids are maybe going to college now and they don’t get that same chance of really understanding analyses and what it’s all about. Afifi: Let’s move on to how you got into USC [University of Southern California] — tell me some memories about that period. Azen: In 1969 when I completed my Ph.D. in biostatistics as your student at UCLA, I left RAND and went to USC. It was Richard Bellman who encouraged me to go to USC, and he hired me in the Department of Bioengineering. And then that opened me up to developing programming where you could look at real-life data and explore biological relationships. Afifi: OK, that was early on. And then you moved on to the USC School of Medicine and started a new program there? Azen: Yes, and I ended up remaining as a professor at USC, now called the Keck School of Medicine, for 46 years, before retiring a year and a half ago, in 2016. The person who was involved with the USC program, Malcom Pike really wanted me to create a program about statistical applications to medical research. And that’s what triggered the whole thing. I learned what others were doing in this field, and that’s where I would also use some of my language skills, where I would learn that this really was the study done in France, and this was the study done in Germany. Then I came up with ideas to push medical research via biostatistical collaboration at USC. I was grateful at this time to be publishing jointly with Malcolm Pike as well, related to his well known work in cancer in women. We published improved statistical methods in analyzing such data. Afifi: And you actually started a new degree program at USC—the biometry program? Azen: Yes, by 1970 I had established biostatistical courses in data analysis, mathematical statistics, and clinical trials, leading to my development of the M.S. degree program in Biometry in 1975, and the Ph.D. program in 1978. It later became the ‘‘biostatistics’’ program. I think I was probably told at that early time that it should have a different name to distinguish it from the UCLA program. Biometry, I remember, dealt more with doing studies with fish, biological processes, and other things, and later came to mean the measurement of tissue or bodily structures, such as the iris. So we felt biostatistics was clearer at that point, and changed the program’s name to that. Duncan Thomas, an excellent biostatistician from McGill University in Canada, came to USC in the early 80’s, and he and I then partnered in running the program. But the reality is that UCLA was very important, thanks to you, because you opened me up to the whole thing as an interesting strategy in biomedicine. And I’m pleased to have been elected to the UCLA School of Public Health Alumni Hall of Fame, even though I spent 46 years of my career at the rival school across town, USC! After starting the MS and Ph.D. programs, in 2001–2003 I also helped create programs in molecular epidemiology and statistical genetics. Then one I’m really proud of helping develop is an M.S. degree in Translational, Clinical and Biomedical Research at USC, geared towards medical professionals who wanted to add a focus and concentration on conducting research. Altogether, these programs train over 200 graduate students every year now. I really enjoy providing mentorship and collaboration to the next generation of students, giving them the opportunity to work together with faculty and develop different career paths. To grab students’ attention, I loved using humor and music in my classes as well — in fact in 1998 I received the USC Associates Award for Excellence in Teaching. 6 Interview / Computational Statistics and Data Analysis 131 (2019) 2–9 Afifi: That’s amazing! Particularly when teaching what can be a dry subject like statistics! OK, so let’s talk a little bit about the journal that you founded, Computational Statistics and Data Analysis — CSDA. Azen: Yes. I was the Founding Editor of the CSDA journal, starting it in 1982. CSDA Associate Editors Joyce Niland and Abdelmonem A. Afifi, Co-Editor Erricos Kontoghiorghes, and Founding Editor Stan Azen, in Pasadena, CA, 2001 Afifi: When it was started, I know that I had the honor of being one of the associate editors of that journal for many years. Azen: Right. And I remained CSDA Editor-in-Chief for 30 years. The journal is the official publication of the IASC [International Association for Statistical Computing], and now I understand it’s also the official publication of the network Computational and Methodological Statistics [CMStatistics], so its reach is continually increasing. Peter Naeve was the first co-editor for the European Region, a section of ISI [International Statistics Institute], which is still mentioned on the cover of each CSDA issue. While initially there was no North American Region per se, there was a connection with the Interface Foundation in the US, so in this way CSDA incorporated North America from the beginning as well. Early on Allmut Hormann created the SSN [Statistical Software Newsletter] within CSDA. When Allmut retired after many years of excellent leadership, my wife, Joyce Niland became the SSN editor. With the growth of the Internet, the SSN was moved to become one of the first online software newsletters. Now that I think about it, it’s interesting that my work on the ARPAnet at RAND years ago was now being put to use to support the journal I founded! Peter Naeve was succeeded by Erricos Kontoghiorghes to represent the European region. Then along the way the journal became so busy that I really needed a co-editor. And even though we’d never met, based on his great work I invited Erricos to become my co-editor, and he really re-invigorated and expanded the scope and reach of CSDA. I remember the first time he and I met, we went up to Santa Barbara, California, where Joyce and I have a second home, and hung out at the beach together. We were so very similar in our passion for the journal, sense of humor, and enjoying watching the pretty girls at the beach! And then when IASC established an Asian Region, Jae C Lee became our co-editor for that area, really enhancing CSDA even further. CSDA now provides global contributions in computational statistics and data analysis, and also provides training for the next generation coming into the field. It now has an impressive impact factor of 1.4. The journal continues to be led very successfully by Erricos, now with co-editors Ana Colubi and Byeong Park as well. And a lot of its success had to do with the fact that wonderful associate editors, such as you, were involved. Afifi: Thank you! Now tell us about your contributions to research in the field. I know that several papers came out of your dissertation on dealing with missing values in longitudinal data. Later on you moved into some other areas, such as ophthalmology and cardiovascular disease? Azen: Yes. My dissertation opened me up to studies that continued along those lines of longitudinal data analysis. I’ve published over 20 papers in statistical methodology, but what really grabbed my interest was collaborating on applied statistics in many different fields of medicine. Ophthalmology was one of the hot areas, and I’d always attend the ophthalmology ARVO [Association for Research in Vision and Ophthalmology] meetings in Florida. Over the years I conducted research and wrote papers on many topics including glaucoma, retinal evaluation in HIV-positive patients, cataracts and macular degeneration. I also coordinated the Los Angeles Latino Eye Study –or LALES – which was the first of its kind Interview / Computational Statistics and Data Analysis 131 (2019) 2–9 7 for Latinos. This study looked at prevalence and risk indicators of visual impairment and blindness in Latinos in Southern California. The cardiovascular disease studies and papers were done over a period of time and focused on many factors of heart disease risk involving fat and cholesterol intake, methods and baseline data from early versus late intervention trial testing. A study on cardiac risk factors and age-related macular degeneration was associated with LALES as well, blending the two disciplines. In 1987, I began working on the Transfusion Safety Study — TSS. My co-director of the TSS data coordinating center was Joyce Niland, who has now been my wonderful wife and life partner for 25 years. For this study, Joyce and I created one of the first microcomputer-based distributed data management systems for a large cooperative study of transfusion-associated AIDS. The study lasted a number of years, and began even before HIV [Human Immunodeficiency Virus] was discovered. Once tests to detect HIV became available, the study was able to estimate HIV prevalence and its relationship to AIDS in ways that no other study could do. Stan Azen and wife Joyce Niland, 2003 Another major area of collaboration was in OT [Occupational Therapy]. I had a joint appointment in this department at USC. A major project for which I was the primary statistician was the ‘‘Well Elderly’’ study, led by one of my early very successful graduate students, Florence Clark, who became chair of the OT Department. In this study retired adults were randomized to either no intervention, just social ‘chit-chat’ interactions, or in the third group, full support from a certified occupational therapist to help them achieve their goals in life — it could be learning to paint, gardening, volunteering, anything. The study was highly successful. This last group was found to be significantly better on quality of life scores, and lower on medical costs going forward. The study was published in JAMA [Journal of the American Medical Association], and shown on the nightly news, with a 2.5 million ‘‘Nielsen’’ rating of viewers who saw the story! I even played piano music for a video on the study, again blending my passions of research and music. And I always used my mother’s success story in her later years as a shining example of what can be achieved. The American Occupational Therapy Association awarded me in 2003 for my participation in this work. Afifi: Fabulous! These studies and others can all be found in your impressive CV, correct? Azen: Yes, correct — I think I’ve published over 400 papers now, along with the two editions of our textbook. Afifi: That’s impressive! So I know that music remains a big part of your life, Stan — both composition and performance and just enjoying it. Tell me a little bit about how music shaped your life. Azen: Well, some of the early things were when I wrote music to present certain themes, like for the OT study. During my years at UCLA I was an active composer, and wrote music for a documentary on autism, that went to the Cannes film festival. I also wrote music for several plays, including the French pataphysical play, ‘Ubu Roi’, which was performed at the Coronet Theater in Los Angeles. In 2000, I was playing piano at a cocktail reception for a meeting on cancer research where my wife Joyce was a speaker. Once I learned how to play jazz and standards, and starting playing professionally at restaurants in LA, she always volunteered me as the cocktail pianist at meetings! We had recently produced my first piano CD, and a representative from a drug company in France heard me playing. He came over and said, ‘‘Would you mind letting us have 10,000 copies of your CD?’’ And the reason they wanted the CDs was to give them out to people being treated with chemotherapy for cancer, so the CD would really help them with relaxation. Soon after this, I got a letter from a patient saying, ‘‘Dear Professor Azen: I’m 80 years old 8 Interview / Computational Statistics and Data Analysis 131 (2019) 2–9 and I have cancer, but I listen to your CD every day and it keeps me alive’’. I have this letter framed on the wall of my home office. Afifi: Fantastic. Azen: And see, these are the fantastic things. Music is helpful in the sense that it gets people away from the stress, and can be soothing and familiar—you know, such as a musical theme. So I did become aware early on that music and doing research were both passions of mine. This is something my wife and I love to do together, she is a real music fan, especially jazz. Over the years, I have recorded quite a bit of music. I now have made 3 piano CDs, all with themes that are statistical ‘‘spinoffs’’ that Joyce and I concocted: ‘‘Regressive Moments’’ [1997], ‘‘Timed Series’’ [2004], and ‘‘Clustering En Stanza’’ [2010]. The phrase ‘En Stanza’ is an anagram on my name ‘Stan Azen’ that Joyce made up! She has served as my music producer, photographer, graphic designer, and percussionist for these CDs. One of my favorite things when I would meet a new graduate student is to ask: ‘‘Do you have a girlfriend/boyfriend? If so I’ll play piano at your wedding, free of charge!’’ And sure enough, I’ve had the honor of playing at the weddings of about 20 of my graduate students — many of the songs I’ve played for them are included on my third CD. Joyce and I love traveling the world together as well. We have been lucky to visit throughout Asia, Australia, Europe, the Middle East, and Africa. I would get to play piano at international meetings we’d attend as well. In about 1995 I played some jazz in Seoul, at the reception to the Satellite meeting to the ISI Beijing meeting. When our dear friend Jae C Lee retired, I was honored to play some music for his retirement celebration in Korea. Joyce and I wrote an anthem together for the IASC, and I performed this song at the conference in Australia — also recorded on CD #3! CSDA Co-Editor Jae C Lee with Stan Azen in Seoul, Korea,1994 Italy is one of our favorite countries — we love to go to Positano on the Amalfi Coast in particular, and I play at every restaurant there that has a piano! I was also very moved when in 2006 I received an Honorary Doctorate in Biomedical Innovations from the University of Salerno, Italy. So science and music keep coinciding throughout my life. Afifi: I personally have benefited from your musical background because you taught me how to play piano. And the first piece that you had me play was from the notebook for Anna Magdalena of Johann Sebastian Bach, and it was the ‘‘Minuet in G’’. And you also selected a piece for me from Prokofiev’s ‘‘Music for Children’’. So let’s talk about the future. What is it you want to do now, Stan? Azen: One answer is, for the future I don’t want to just sit around and do nothing. So what I’ve been doing since I’m now retired is to come up with lists of homes for senior citizens, where I can play piano to groups of people that are living in the retirement home. And I try to play songs from the past that these people would know and can sing with me, and then that makes them loosen up and enjoy the music. And that has been happening — I start playing, and some 80 year old lady comes over, takes my copy of the music and starts singing it, and she says: ‘‘Oh, this is perfect! I feel a thousand times better!’’ Afifi: Wonderful. Azen: And I’ve been doing that now maybe two or three times a week. And every time, people keep coming over and hugging me — and they’re not Huguenots! Bad joke number 5. Afifi: Hugging and Huguenots — I got it! Very nice. Azen: Oh, and at a gala last April, I was honored by USC as a 2017 recipient of the Faculty Lifetime Achievement Award, which you attended. This is a card, by the way, that one of my students sent to me. It says, ‘‘Hi Dad, congratulations on your lifetime achievement award. Well, it’s about time you were honored with this award. I think your accomplishments and contributions merit about ten lifetime awards. No matter what the achievement, we can’t ever discount one of your greatest—the impact you’ve had on the lifetimes of your students. Thank you again, Dad, for lifting me up both professionally and personally at a time when I really needed it. Thank you and congratulations’’. Interview / Computational Statistics and Data Analysis 131 (2019) 2–9 9 Stan Azen, with A. A. Afifi, displaying his USC Lifetime Achievement Award, 2017 Afifi: This is from one of your former graduate students? Azen: Yes. Afifi: And she calls you Dad? Azen: She calls me Dad, yes. Afifi: She calls you Dad. Wow. [A full copy of Dr. Azen’s CV, and links to his piano CDs, can be found at StanleyAzen.com] Abdelmonem A. Afifi UCLA Fielding School of Public Health, Biostatistics, Mail Code 177220, Los Angeles, CA 90095-1772, United States E-mail address: afifi@ucla.edu. Received 28 January 2018 Received in revised form 28 July 2018 Accepted 26 August 2018 Available online 11 October 2018