As we reported in the previous issue, the 2016 George Box Medal has been awarded to one of our contributing editors, David Hand. This medal is awarded by the European Network for Business and Industrial Statistics for outstanding contributions to industrial statistics (see http://www.enbis.org/awards/george_box_medal/?_ts=18012)

David Hand

David Hand is well-known for his work in the personal banking and credit scoring domain, but has also worked extensively in the pharmaceutical and other sectors.

David Hand was Professor and Head of the Statistics Section from 1999 to 2010 at Imperial College London, where he is now Emeritus Professor of Mathematics and Senior Research Investigator. He is also Chief Scientific Advisor to Winton Capital Management, on the Board of the UK Statistics Authority, and chair of the Board of the UK’s Administrative Data Research Network. He previously served two terms of office as President of the Royal Statistical Society—the first person to serve twice since Lord George Hamilton in 1915.

David’s work in industry has been previously recognized, for example, by 2012 the Credit Collections and Risk Award for contributions to the credit industry—the first time that award had been made to an academic. The work of David and his group figured as an impact factor case study in the UK’s most recent Research Excellence Framework. In 2013 he was made OBE by the Queen, for contributions “to research and innovation”.

He has written over 300 scientific papers and published 29 books. His first book, Discrimination and Classification, appeared in 1981, a topic which continues to fascinate him. As the world rolls on and computer power advances, so new challenges of classification continue to arise, such as high dimensional problems in genomics, and streaming data classification problems using web data.

More recent books include Principles of Data Mining; Measurement Theory and Practice: The World Through Quantification; Information Generation: How Data Rule Our World; The Wellbeing of Nations: Meaning, Motive, and Measurement; and Measurement: A Very Short Introduction. His popular book The Improbability Principle: Why Coincidences, Miracles and Rare Events Happen Every Day, attracted a lot of attention—and many radio and television appearances.

David says that the range of his interests, from how to squeeze answers from data through to fundamental issues of what data actually are, is apparent in the range of these titles: from data mining to measurement, from applications to basic theory. In classification, for example, papers which have attracted particular interest include “Classifier technology and the illusion of progress, showing and presenting reasons why progress may be less rapid than it appears,” and “Statistical fraud detection: a review,” both of which appeared in Statistical Science, as well as a series of papers demonstrating that the very widely used measure of classifier performance—the area under the ROC curve—had a fundamental conceptual flaw. This, he says, was a very exciting discovery, since it showed that careful thought about even well-established concepts and methods can reveal unexpected properties.

While he is an enthusiast for data, he has the statistician’s natural caution, and urges people to think twice before getting carried away by uncritical enthusiasm for big data. “Big promise, yes. But big challenges, too.”

He served as joint editor of the Royal Statistical Society journal Applied Statistics and established the journal Statistics and Computing, serving as its Editor-in-chief for 11 years.

He believes that statistical research should be primarily driven by practical questions that matter—by problems that people want solved. And, he says, the history of the development of statistics shows that this has in large part been the case, with new methodological developments arising because of the challenges of new application domains and their new kinds of data.

He refuses to be pinned down as belonging to one of the many schools of statistical inference, believing that one should choose the tool which is most appropriate for the job.

He says that his view of statistics is summed up in the way he began his book Statistics: A Very Short Introduction: at last, he says, with the big data and data science revolution, people are beginning to understand what we have been telling them for decades, namely that “Statistics is the most exciting of disciplines”.