"The reason I do statistics is I’m interested in so many different things": An interview with Peter Guttorp


  • Author: Alison Oliver, Statistics Views
  • Date: 30 Jan 2019
  • Copyright: Image appears courtesy of Professor Guttorp

Peter Guttorp is Professor Emeritus of Statistics at University of Washington, Professor at the Norwegian Computing Center, and Adjunct Professor of Statistics at Simon Fraser University. He obtained a degree in journalism from the Stockholm School of Journalism in 1969, a B.S. in mathematics, mathematical statistics, and musicology from Lund University, Sweden, in 1974, a Ph.D. in Statistics from the University of California at Berkeley in 1980, and a Tech.D. h.c. from Lund University in 2009. He joined the University of Washington faculty in September 1980 and retired in June 2015.

Dr. Guttorp's research interests include uses of stochastic models in scientific applications in hydrology, atmospheric science, geophysics, environmental science, and hematology. He is a fellow of the American Statistical Association and the Institute for Mathematical Statistics, and an elected member of the International Statistical Institute, for which he is a Vice President 2017-2021. During 2004-2005 he was the Environmental Research Professor of the Swedish Institute of Graduate Engineers, and in 2014 he was a Chalmers Jubilee Professor.

Alison Oliver talks to Professor Guttorp about his career in statistics so far.

thumbnail image:

1. You obtained your first degree in journalism from Stockholm School of Journalism before moving on to mathematical statistics at Lund University and later a PhD in statistics at Berkeley. What inspired this move?

I was a freelance journalist in Stockholm for a long time, and with the Swedish system, it turned out to be easy to do part-time study, because you could obtain loans and that helped you financially in a dry spell. I applied for a scholarship to go to Berkeley in order to take courses in statistics and the natural sciences, wishing to become a science journalist. Jerzy Neyman taught one quarter in Biology, one quarter in Hydrology, and one quarter in Astronomy. That was a lot of fun. I ended up completing a PhD at Berkeley. I got a job and started a family, and suddenly I was a statistician. There are lots of similarities; I mean the kind of articles I would have written were in-depth articles where you spent a couple months getting the background straight, and instead I now take a couple of years to get the background straight. But it’s the same idea. You get to dabble in all sorts of interesting areas. There are similarities between being a statistician and being a journalist.

2. Your research interests include uses of stochastic models in scientific applications in hydrology, atmospheric science, geophysics, environmental science, and hematology. What are you working on currently?

I’m working on adaptation to sea-level rise with authorities in Denmark. Two students and I are also working on large earthquakes. Every 500 years or so we get an earthquake, of a large magnitude, the kind that don’t happen that often, but when they happen you need to be prepared, and we are not. We have a large multidisciplinary group: us, engineering, working with the local rescue people, and people who do the planning for emergencies. My group also does paleo seismology and modern seismology, we have a lot of different things that are going on. It’s good to be multidisciplinary. I think the important thing about it is really that we have the local authorities involved. In the US right now, the local authorities are who you need to talk to, because the federal authorities aren’t interested.

3. You are currently Vice President of the International Statistical Institute. What are your objectives for this time ahead?

There are a few things I want to work on. The most important one is to get us more involved in international science policy. The ISI used to be an affiliate member of the Council of Scientific Organizations—it’s something different now—anyway, it’s the group that the United Nations goes to when there’s an international program that’s going to be set up. Like, the IPCC was formed through that. Since we weren’t members, there were no statisticians involved in the IPCC, and it was hard to get more understanding of the importance of statistics. It’s getting much better. We had a statistician on the scientific board the last round, so that definitely helps.

4. You were also President of the International Environmetric Society. What are your memories when you look back on your Presidency and what do you feel were your main achievements?

The main thing I worked on my entire presidency, and it took a long time to be finished, was becoming affiliated with ISI. At the same time the ISI was starting to think about an environmental section, and I, along with many, didn’t think there were enough environmetricians around to have two international organizations. That eventually worked out, but it took a lot of effort.

5. You were Editor of the journal Environmetrics. What can the journal still offer to this day in the competitive market that makes it stand out from its competitors?

It’s the best environmental statistics journal. I think the work that Walt has done has brought it up to a much higher level. I don’t care much for publication statistics, but you can look at them for trends, and it shows up there that it’s a more-cited journal, it’s a better journal.

6. What do you think have been the most important recent developments in the field and will these influence your teaching in future years?

Well, I mean, the biggest influence on the field has been borrowing ideas from statistical physics. That has allowed us to analyze problems that we wouldn’t know how to do. So when I was a grad student, I’d think, “Oh yeah, one ought to be a Bayesian. But of course, I can’t do any of the problems in that framework.” But now you can! So I think that has been a fundamental change.

I’ve been working on big data for a fairly long time. Climate models are the order of magnitude of the purchase patterns of moderately large online stores. These are huge things that you need to handle appropriately. One could call it computer science, but it isn’t; it’s really computer applications that have become necessary for statisticians to understand.

One example is: there’s a telescope that’s going to come up in somewhere in South America in the next few years, and the idea is they’re going to take two galaxy surveys a week, looking for supernovas, etc. A supernova will last for about two weeks, so if you want people to observe it you need to discover it. You have a humongous dataset that’s collected twice a week, and it has to be transported so that you can analyze it, and you have to compare what you see today to what you’ve seen in the past. You can’t compare pixel by pixel. We don’t have computers that are that fast. We have to therefore come up with really new ways of thinking about the problem. I mean that’s what big data has done: statistics used to be about squeezing maximal information out of a tiny dataset. Now, with this, we can get minimal amounts of information from a huge dataset. You have to throw away information, because you can’t keep up. It’s interesting as it’s changing the way you attack problems.

7. Your research has been published in many journals and books: is there a particular article or book that you are most proud of?

I think my stochastic modelling book, because it’s one of the few books that does both statistics and applied probability and has datasets. There are lots of applied probability books that don’t do statistics, and there are a few statistics books on stochastic processes that don’t have data, and there aren’t very many like mine.

8. What is the best book in statistics that you have ever read?

I think the book I come back to the most is Bertil Matérn’s thesis. It has a lot of the fundamentals of spatial statistics. A lot of the tools that we use today, we can find the theory there. It’s stuff that he knew in the early ‘60s. It got republished by Springer. I think that edition has sold out, too. It needs to be republished again, in the classics series.

9. What would you recommend to young people who want to start a career in statistics?

My advisor told me when I started: don’t use data until you’ve obtained tenure. That was very good advice that I did not follow.

Follow your heart. Do stuff that you love to do. The reason I do statistics is I’m interested in so many different things. I’ve had NSF funding to learn planetary science. That ended up in papers about the pressure cycles on Mars. It’s great fun! I’ve done hematology for many years. It’s moved that field from being a field of differential equations that didn’t fit to stochastic models that does a much better job of fitting, but there’s an educational aspect there in that we have to teach people how to think about these things. There’s a tendency in biology to think of “random” as meaning “independent”. It’s not.

10. Who are the people who have been influential in your career?

Jerzy Neyman who was my master’s thesis advisor and stopped talking to me for a couple of years after I didn’t get the result he expected. He got over that. David Brillinger. David Cox. Those are probably the three most influential. Georg Lindgren to a large extent. We work together now, but he was one of my first teachers.

Related Topics

Related Publications

Related Content

Site Footer


This website is provided by John Wiley & Sons Limited, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ (Company No: 00641132, VAT No: 376766987)

Published features on StatisticsViews.com are checked for statistical accuracy by a panel from the European Network for Business and Industrial Statistics (ENBIS)   to whom Wiley and StatisticsViews.com express their gratitude. This panel are: Ron Kenett, David Steinberg, Shirley Coleman, Irena Ograjenšek, Fabrizio Ruggeri, Rainer Göb, Philippe Castagliola, Xavier Tort-Martorell, Bart De Ketelaere, Antonio Pievatolo, Martina Vandebroek, Lance Mitchell, Gilbert Saporta, Helmut Waldl and Stelios Psarakis.