“Pay attention to the real data which truly motivates the development of statistics”: An interview with Professor Guang Cheng, winner of the Gottfried E. Noether Young Scholar Award

Features

  • Author: Statistics Views
  • Date: 08 Feb 2013

Further to last week's feature on the winner of the Gottfried E. Noether Senior Scholar Award, Professor Joseph Gastwirth, at last year’s Joint Statistical Meeting in San Diego, the Gottfried E. Noether Young Scholar Award was won by Guang Cheng, Assistant Professor of Statistics at Purdue University for ‘outstanding early career contributions to nonparametric statistics ‘(ASA, 1st October 2012).

Statistics Views talks to Professor Cheng about his background and his research on semiparametric efficiency, bootstrap theory and empirical processes which will last a lifetime.

thumbnail image: “Pay attention to the real data which truly motivates the development of statistics”: An interview with Professor Guang Cheng, winner of the Gottfried E. Noether Young Scholar Award

1. Congratulations on winning the Gottfried E. Noether Young Scholar Award for 2012. With the teaching career that you are continuing to enjoy, what was it in the first place that inspired you to pursue a career in statistics and economics? When and how did you first become aware of statistics as a discipline?

Interestingly, there is no statistics department at Tsinghua University where I got my undergraduate degree. Hence, I learned most of my statistical knowledge from one Econometrics course offered by the University’s Department of Economics and one Mathematical Statistics course, offered by the Department of Applied Mathematics whilst an undergraduate in Economics. Later on I received very rigorous statistical training during my graduate school at University of Wisconsin - Madison. My first experience with statistics was as a powerful quantitative tool for analyzing economics data, but then realized it was an independent discipline. This might be a bit different from my peers, but it provides me with a rather different perspective on statistics. I like statistics because we can see an immediate influence of statistical theory on our understanding of data in the sense that it establishes the concept of optimality/efficiency or provides deep insights. In other words, the usefulness of good statistical theories is “verifiable” in practice.

2. What are your main areas of interest in statistics and why?

My general research area is Mathematical Statistics, in particular Semiparametric Statistics, Empirical Processes and Bootstrap. Recently, I have become interested in the joint asymptotics and inferences for semi-nonparametric models where parametric and nonparametric components are both of interest. This is a new asymptotic framework beyond the classical marginal asymptotics. I am particularly curious about the interplay/connection between two different optimality concepts: semiparametric efficiency and nonparametric minimaxity. In my opinion, the power of these theories will be realized when the statistical models become more and more complicated, e.g., its application to semi-nonparametric model selection. Obviously, what I propose here will be a life-long endeavour for me. 

I like statistics because we can see an immediate influence of statistical theory on our understanding of data.

3. Do you think that statistics undergraduates and postgraduates starting out today are under more pressure to publish and to obtain grants than when you were a student yourself? Does this in turn affect the teaching of statistics?

Yes, this can be easily quantified by counting the increasing number of publications in the recent and current job applicants’ CVs, namely publication inflation. Every search committee experience surprises me a bit. However, I do feel that less pressure might be better for the fresh PhDs, who are in their most creative and energetic period. They need to pursue the areas they feel most interested in/most important rather than work on the areas which easily produce more papers, e.g., their thesis directions. Meanwhile, the greater pressure to publish results in the following teaching/mentoring challenge: how do we convince graduate students to commit themselves to work on the difficult mathematical/statistical theory courses, (e.g., measure theory and empirical processes), if their publication record is affected early in their career?

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

Pay attention to the real data which truly motivates the development of statistics.

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

My research ideas continue to come from reading the most recent literature, e.g., from Arxiv, or communicating with the experts who have much better vision than me. I usually incorporate the most recent results from my research area into a graduate level special topic course. The statistics students at Purdue really appreciate learning at the cutting-edge of research.

The students really appreciate learning at the cutting-edge of research.

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

The joint asymptotics and inferences I am working on currently are what excite me so far. My plan is to invest a significant amount of my research time in this area in the future.

7. 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?

In general, I believe that it is about how to understand new types of data (e.g., high dimensional sparse data, massive data and network data) in a statistical way. In my humble opinion, the following areas might need more attention in the next few years. First, it seems to me that there is a long way to go for before we lay down any type of rigorous theoretical foundation for analyzing high dimensional data. In fact, it could be very different from the existing theories. For example, the concept of semiparametric efficiency is certainly not well defined in the high dimensional case. A further direction is that we need to develop computationally efficient bootstrap procedures for massive data by taking advantages of modern parallel computing approaches. Finally, I also feel that valid post-model selection inference is not well enough understood yet, even though we have seen some very exciting ideas recently.

We need to develop computationally efficient bootstrap procedures for massive data by taking advantages of modern parallel computing approaches. Finally, I also feel that valid post-model selection inference is not well enough understood yet. 

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

There are certainly quite a few challenges as our profession is growing. One challenge is how to effectively establish a research collaboration platform within the statistics department or cross different disciplines (with statistics as a core component). The Focused Research Group sponsored by the National Science Foundation seems to be a valuable effort towards this direction. Another challenge could be the incorporation of big data and computing into the graduate and undergraduate curriculum. This is essentially one of the teaching emphases at Purdue Statistics.

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

In the research part, I would like to mention my former PhD advisor Professor Michael Kosorok and my academic role model Professor Jon Wellner. They continue to teach me how to think deeply and independently. With respect to mentoring, I really appreciate my colleagues Professor Jayanta Ghosh and Professor Rebecca Doerge. And, finally I would like to acknowledge that Purdue Statistics and National Science Foundation are very generous in supporting my career.

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.