History


AT&T has a long history in statistics, dating back to Shewhart's invention of control charts in 1924 and Dodge and Romig's development of double-sampling inspection plans in the 1930's. One of Shewhart's interests was what is now called the robustness of statistical techniques.

After World War II, John Tukey embarked on his dual career at AT&T Bell Labs and Princeton University, carrying out his pioneering work in spectrum analysis, the Fast Fourier Transform, multiple comparisons, exploratory data analysis, and a host of other topics in statistics. The first person hired as a statistical researcher (in 1952) was M.E.Terry.

Walter A. Shewhart John Tukey M. E. Terry

Starting in 1959 the Statistics Research departments were greatly expanded. Among the early recruits were M. B. Wilk, Ram Gnanadesikan, and Colin Mallows.

M. B. Wilk R. Gnanadesikan C. L. Mallows

In addition to the full-time staff members, many distinguished statisticians visited over the years, including D.R.Brillinger, F.J.Anscombe, and D.R.Cox.

In 1984, with the government imposed break-up of the Bell System, about 1/3 of the group went to Bellcore. In the years following, there were further divisions of the company, but these later splits were self-imposed. In 1997, AT&T spun off Lucent Technologies as a maker of equipment and technology. Some of those functions have since been spun off into other companies: Avaya and Agere. Lucent kept rights to the Bell Labs name. The part that retained the AT&T brand is the telecommunications carrier, and network designer and operator; it has since been purchased by S.B.C, one of the so-called Baby Bells, but has retained the AT&T name.

Beginning in the 1970s, our group changed the field of statistical computing with the development of the S programming language by John Chambers Rick Becker, and Allan Wilks. It was eventually licensed as S-Plus and later spawned the open source statistical language R. S/S-Plus/R has become the de-facto statistical programming language for many diverse disciplines and its flexibility, robustness and ease of use is a credit to its designers.

John Chambers Rick Becker Allan Wilks

Another success story was in fraud detection, where a flurry of projects spearheaded by Daryl Pregibon established AT&T as the industry leader in fraud modelling using statistical profiling. Through this work, our department played a pioneering role in the explosion of the data mining field in the late 90's and early '00s.

Daryl Pregibon

A guiding principle in all the work at AT&T Research has been the need to keep theory and practice in touch with reality. A common recurring theme has been an emphasis on graphical methods of display and analysis of data, and continues on in our work on data mining, data visualization, and interactive data analysis.

The demand for statistics at AT&T Labs is strong. Our tradition of doing world-class, fundamental research in statistics goes back to people like Walter Shewhart and John Tukey. AT&T Labs wants to continue this tradition by maintaining a strong research presence in statistics.