“Statistics provides a challenge somewhat akin to Sherlock Holmes’ task: how to find hidden truth in any data, from small to big.” An interview with Samuel S. Wilks Award winner Kanti Mardia

Features

  • Author: Statistics Views
  • Date: 04 Feb 2014
  • Copyright: Image appears courtesy of the University of Leeds

At last year’s Joint Statistical Meetings in Montreal, Canada, Professor Kanti V. Mardia was awarded the Samuel S. Wilks Award for ‘extensive work covering a wide span of applied and theoretical research, including seminal results in shape analysis, spatial statistics, multivariate analysis, directional data analysis and bioinformatics with special applications to geostatistics, image analysis and protein structure; for the international dissemination of statistical thought and innovative ideas through research publications, presentations, books, monographs, the establishment and running of annual research workshops and interdisciplinary centers; and for his insightful guidance of future generations of statisticians.’

Professor Mardia obtained his B.Sc. degree in mathematics from Ismail Yusuf College (University of Bombay) before going on to study for M.Sc. degrees in statistics and pure mathematics from Universities of Bombay and Poona respectively, and the PhDs in statistics from the University of Rajasthan and the University of Newcastle. For significant contributions in statistics, he was later awarded a D.Sc. degree by the University of Newcastle.

Professor Mardia is currently Senior Research Professor, a position which was specially created by the University of Leeds after the expiry of his position in 2000 as Chair of Applied Statistics which he had held since 1973. He is also a long-term Visiting Professor in Oxford and Adjunct Faculty in the renowned Indian Business Institute (IIMA), Ahmedabad. He was a founding Director of CoMIR (Centre of Medical Imaging Research). Due to his initiative and efforts, the Centre of Statistical Bioinformatics (CoSB), Leeds, has been recently established.

His publications comprise over three hundred articles in statistical journals and numerous books, including two for Wiley, Statistical Shape Analysis and Directional Statistics. He was awarded the Guy Medal in Silver (2003) by the Royal Statistical Society.

Here, Statistics Views talks to Professor Mardia about winning the Samuel S. Wilks Award and the research and career that has led to this honour.

thumbnail image: “Statistics provides a challenge somewhat akin to Sherlock Holmes’ task: how to find hidden truth in any data, from small to big.” An interview with Samuel S. Wilks Award winner Kanti Mardia

1. Congratulations on winning the Samuel S. Wilks Award. Please could you tell us more about your research that led to this award?

It is difficult to pin-point, but my guess is that it is for the cumulative research contribution with possibly, my particular brand of leadership in interdisciplinary research, with taking Leeds and the UK forward in geosciences, image analysis and bioinformatics; in particular I started working when these subjects were just emerging, so the Leeds Annual Statistical Research (LASR) workshops helped in crossing the research boundaries at the right time.

2. With an extensive educational background in statistics, when and how did you first become aware of statistics as a discipline and what led you to pursue a career within the subject area?

In my first degree from Bombay University I had the option either to take astronomy or statistics under mathematics major. There was no teacher to teach astronomy so I ended up by default taking statistics as one of the modules. Subsequently, since I passed with flying colours in my degree I had a lot of choice and at that time it was very competitive to get entry into the MSc in Statistics; thus I ended with a specialism in statistics and consequently my career was bound with statistics.

3. In terms of your research, what are you currently focussing on? What are your main objectives and what do you hope to achieve?

I am currently focussing on protein structure and functions which has applications in drug discovery, medicine and evolutionary biology. Proteins are the workhorses of living systems so these are extremely important, but there is still a lot of mystery in their function. When they malfunction, it leads to various diseases, such as Altzheimer’s, cancer, and so on.

I am collaborating with a group in Copenhagen and another group in Oxford University, and I hope this will contribute to the challenge of working in this area. One of the hardest problems in biology is that of protein folding, which means how protein from amino acid sequences falls into 3 dimensions. What I am working on is, in a tangential way, solving this puzzle.

I am currently focussing on protein structure and functions which has applications in drug discovery, medicine and evolutionary biology... One of the hardest problems in biology is that of protein folding, which means how protein from amino acid sequences falls into 3 dimensions. What I am working on is, in a tangential way, solving this puzzle.

4. You have authored several publications including two books for Wiley, Statistical Shape Analysis and Directional Statistics, and several articles for Wiley journals such as Biometrics and the Canadian Journal of Statistics. Is there a particular article or book that you are most proud of?

This question reminds me of my correspondence with Benoit Mandelbrot, who invented fractals. When I asked him to send one of his best articles with his signature, for the birthday of my son, Dr Hemant Mardia, he sent me his latest article, saying that as it was the latest, it was therefore the best. This echoes my thoughts. Just now I wrote a long article in the Journal of the Royal Statistical Society Series C, also published by Wiley (Mardia, K.V. 2013. Statistical approaches to three key challenges in protein structural bioinformatics, J. Roy. Statist. Soc. C, 62, 487-514). I am proud of that article, which I wanted to be a discussion paper for the Society, but unfortunately the paper did not make it as such because perhaps those who could referee it happened to be mostly my collaborators, so the editors had problems in getting it refereed for a discussion paper! There are many papers I am proud of, including my discussion papers in the JRSS B of 1975 and 2001. I am also proud of my three books, Multivariate Analysis, Directional Statistics and Statistical Shape Analysis; the last 2 particularly as they unified treatment of an emerging area. Multivariate Analysis, although written in 1979, still sells, so has turned out to have a good lifespan in spite of many good new books coming out in this area. In fact, the second edition will be published with Wiley soon.

5. What will be your next book-length undertaking?

I would like to complete the book on Spatial Analysis (with John Kent) which I first started in 1979 (John joined me in 1986 or so), and I am hoping it will be completed this year, again with Wiley.

6. In January 2007, you met the President of India, His Excellency Dr. Abdul Kalam, and discussed with him the topic of “Statistics, Science and Spirituality”. Could you please tell us more about this talk and its outcomes?

Thanks for asking this. Someone arranged a brief interview with him for 5 minutes; however, our discussion went on for half an hour. One topic was the axiomatic postulates that I have produced for Jainism (Jainism influenced Gandhi in his principle of non-violence; it is claimed to be a pre-Vedic religion). Dr Kalam knew about the Jain philosophy but was not aware of my work. Also we discussed how Jain ideas foreshadowed statistics in relation to the principle of inference from samples to population, the idea of meta-analysis (anekantvad - many-sided view), non-absolutism in scientific discovery (syadval). He himself is a scientist, but he is also interested in philosophy; in fact subsequently he has written a book with a very important Jain guru (the late Mahapragna). We both believe that divine values are necessary in science and religion. Einstein summarised this beautifully: ‘Science without Religion is lame, Religion without Science is blind’.

7. Do you have any advice for students considering a university degree in statistics?

Uncertainty is everywhere: information technology is giving us a lot of data and therefore statistics provides a challenge somewhat akin to Sherlock Holmes’ task: how to find hidden truth in any data, from small to big.

8. Over the years, how has your teaching, consulting, and research motivated and influenced each other? Do you continue to get research ideas and incorporate your ideas into your teaching? Where do you get inspiration for your research projects and books?

I have not taught now for about 15 years, except for PhD students. Before then, as Head of Department my attitude was taken from Lord Edward Boyle, the Vice-Chancellor here when I joined Leeds University, whose maxim was to do teaching in the atmosphere of research so they fed each other, especially at MSc level. Most of my research projects come from interdisciplinary projects, and the Leeds Annual Statistical Research (LASR) workshops have been a great platform in which to have interdisciplinary dialogue, and to keep up with the changing scene.

9. What is the best book on statistics that you have ever read?

I have been influenced by the books by Harold Cramer and CR Rao in my earlier career.

10. What do you think the most important recent developments in the field have been? What do you think will be the most exciting and productive areas of research in statistics during the next few years?

I think the most important development has been the computational power which has changed attitudes from small sample statistics to large-scale statistics. What was not feasible before as a practical methodology has now become a reality. Thus there is a significant rise of Bayesian methods. I think the next few years will still see the impact of new data coming from new technology.

The greatest challenge for statisticians is to be ‘ahead’ of computer scientists, that is, to have computational skills, but combined with sound statistical techniques.

11. What do you see as the greatest challenges facing the profession of statistics in the coming years?

The greatest challenge for statisticians is to be ‘ahead’ of computer scientists, that is, to have computational skills, but combined with sound statistical techniques.

12. Are there people or events that have been influential in your career?

The area I come from has many Jain temples with extraordinary geometrical shapes in the architecture, so geometry-driven statistics turned out to be a continual thread in my career. I came to know of my special talent in mathematics at about 9 years old through a teacher in primary school. This teacher, in addition to regular mathematical teaching, would give a mathematical challenge in the form of a puzzle, so if one solved it one was given a cash prize, and I was always first to solve it and receive the prize. This was the first hint of my comparatively special talent in a school in a very small town. Subsequently, I was influenced in my college days by Professor Phadke who took a major part of the teaching himself since this was the first year of running the degree course in Mathematics in Ismail Yusuf College–University of Bombay. I was always the first to get the answer although the questions were never straightforward or obvious. Again he had an interactive style, and in that first batch of 8 students I got first class with distinction, one got third class and the rest failed. When the results came he told me I had saved the Department because otherwise it would have been closed. I still have the Hindi-English dictionary he gave me in recognition. My teaching style has always been interactive. So these are the two formative influences.

In terms of those who have also played a key role in my life, I would say that in my college days it was my eldest brother, and after marriage my wife, Mrs Pavan Mardia. Without her support and encouragement, I am sure nothing much would have happened.

Related Topics

Related Publications

Related Content

Site Footer

Address:

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.