His scientific interests are broad and include methodology, theory, and applications. He has worked in engineering statistics, reliability and degradation modeling, network tomography, design and analysis of experiments (including applications in behavioral intervention research), and quality improvement.

StatisticsViews.com talked to Professor Nair during the Joint Statistical Meetings 2013 in Montreal, Canada where he gave the ASA Deming Lecture. Professor Nair discussed his hopes and objectives for his Presidency of the Institute, how the Institute has developed over the years and responded to the changing needs of the statistical community, what inspired him to start a career in statistics, the theme of his lecture at JSM and his collaborative research methods.

1. Your research interests include methodology, theory and applications. What are you focussing on currently and what do you hope to achieve through your research?

My own research has been very eclectic and that goes back to the way my colleagues worked at Bell Labs. My colleagues were problem-solvers and did not admit to any expertise or to a dedicated interest in one area. We would solve interesting problems, write our papers, develop methodologies and how to implement those methodologies in practice and that is still the way I operate now. The positive aspect of this is that I get to work on so many interesting problems but on the other hands, you can become a jack-of-all-trades so that you do not get your name recognised in one particular area. But this is fine for me and I enjoy my work. At the moment, one of the areas I’m working on in large-scale computational modeling. A lot of the models available these days are so complex that you are unable to observe the data, such as climate modelling or nuclear testing. People are using computer models to design products and processes and so instead of physical measurements, you obtain only assimilated data and then how does one then use this data and approximate those models are interesting problems for me.

I am also working on reliability and risk analysis; credit card fraud, behavioural models; design of experiments, so as you can see, I work on a number of different areas and I like when a colleague and I spot something that is fascinating and we agreed to get together and work on it.

2. Your lecture here at JSM 2013 was the ASA Deming Lecture. Could you please tell us about your theme for the lecture and what points you hope to bring across?

The title was ‘Industrial Statistics: Research and Practice.’ I tried to talk about the historical overview of what has happened in the area of industrial statistics. I started off with the more traditional methods of manufacturing where methodologies used are process control, design of experiments and reliability, and then I moved onto the more recent directions, such as what is happening with Big Data, advanced manufacturing, hi-tech applications, IT, banking finance and fraud detection.

People are not keeping up with the more advanced techniques. Granted, research is always forward-looking so there has to be some gap between research and gap. I talked about how this gap could be reduced (at the ASA Deming Lecture during JSM 2013)

I also talked about the disconnection in research of practice. On the one hand, people are publishing lots of papers on the refinement of methodology but on the other hand, some of the practices are still lagging behind and people are still using very old-fashioned techniques. They are not keeping up with the more advanced techniques. Granted, research is always forward-looking so there has to be some gap between research and practice. I talked about how this gap could be reduced and that was the main theme of my talk.

3. What has been the most exciting development that you have worked on in statistics during your career?

The one area I spent considerable time on understanding a process when I was at Bell Labs was about semiconductor fabrication and trying to develop new methodology to improve the process of fabrication but develop new statistical methods. Not only did we publish papers on our findings but I won a National Science Foundation research grant.

4. 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 exciting development is the so-called ‘data revolution’. We are inundated with data and it is now so easy to collect data and it is much cheaper now to do so. Everything is being measured what whatever you buy in a store to its backing transaction and your purchase then becomes part of the store’s information about their customers. So how do we make sense of Big Data? People talk about how we can use data to solve the world’s problems and I think that is too simplistic a view because a) there is a lot of data but it is not necessarily the right kind of data as you are only collecting what is available and b) the feel of data science which we are working on at Michigan – data retrieval, visualization, if we want to train the next generation of students, we have to be able to have people who can handle the programming as well as the data analysis because right now we have computer scientists who are not good at analysing data and statisticians who are not good at collecting the data, so we need to bridge that gap for the next generation.

A lot of money is being invested in Big Data but there is a lot of interesting research issue for statisticians to solve together with computer scientists on applications for computer science, and for business and industry too.

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

The difficulty with the profession of statistics is the name, unfortunately. It comes from the word ‘state’, meaning the collection and analysis of data from government. When people associate statistics with collecting numbers from government or sports, there is an assumption that we just collect the numbers and crunch them. They do not realise that a lot of exciting work is involved in collecting data and transforming the data into information.

With the emergence of Big Data, this is pushing computer scientists to the forefront…I believe we are losing out on this excitement…so we really have to be at the forefront of this data revolution and this is my main hope.

I have met people whom when they asked me what I do for a living respond with “Statistics was the worst class I ever had.” We have to change our image.

Also, with the emergence of Big Data, this is pushing computer scientists to the forefront and they use more exciting words than we do like ‘data mining’ and ‘machine learning’ so I believe that we are losing out on this excitement and losing our initiative to other people, so we really have to be at the forefront of this data revolution and this is my main hope.

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

For me, I was very fortunate to have a very solid theoretical background but I was also very fortunate to go to Bell Labs and understand how statistics was being used. Senior colleagues there influenced my research career but I really then understood how statistics is a very active discipline and how it is used in real life, and that has really stayed with me throughout my career and really affected how I think about things.

But looking before then when I started off in Malaysia, there was only one university in what was an under-developed country and I had some really good teachers who were very influential in shaping my views and I still remember all of them. Sadly, some of them are not with us anymore. I come from a very poor family and I was fortunate enough to obtain a scholarship to come overseas and I obtained my degree in the US and met my wife there and meeting her shaped my career as I decided not to return to Malaysia. A lot of people stayed there and have not been so lucky. I have been a very lucky and fortunate man.