Session Slot: 2:00- 3:50 Monday
Estimated Audience Size:
AudioVisual Request: Two Overheads
Session Title: Neyman Lecture
Theme Session: No
Applied Session: Yes
Session Organizer: Lindsay, Bruce The Pennsylvania State University
Address: 422 Thomas Building, Department of Statistics, University Park, PA16802
Fax: (814) 863-7114
Session Timing: 110 minutes total (Sorry about format):
Opening Remarks by Chair - 5 First Speaker - 90 minutes Floor Discussion - 10 minutes
Session Chair: Markatou, Marianthi Columbia University
Address: Department of Statistics, 615 Mathematics Building, New York, NY 10027
Phone: (212) 854 3969
1. Computing with Data: Concepts and Challenges
Chambers, John, Bell Labs, Lucent Technologies
Address: Rm. 2C-282 Bell Laboratories Lucent Technologies 700 Mountain Av. Murray Hill, NJ 07974-0636
Phone: 908 582-2681
Fax: 908 582-3340
Abstract: Our use of computing, as statisticians or members of related professions, takes place in a broader context of activities, in which data is acquired, managed, and processed for any of a great variety of purposes. The underlying purposes are diverse: scientific experimentation, business transactions, or managing society itself (the original meaning of statistics). The activities, of which traditional statistical computing is an important but quantitatively small part, form what I call computing with data .
This talk examines how work in statistical software and related areas can contribute to the activities, and in particular, to science. Such a focus is an appropriate topic for the Neyman lecture, since our ability to compute with data is often the limiting factor in applying statistical techniques to science today.
Discussion of what has been done, and is being done now, in statistical software is presented around three basic concepts: language , the ability to express our ideas; objects , the structure we can impose on data; and interfaces , the means of communication within software and between software and its users. To complement the discussion of concepts, we need to look at the challenges to be faced. Here, one challenge subsumes many of the most important ones: the scope of our participation in the larger context. Simply put, we need to greatly enlarge the scope of our activities and of our model for our own rôle in computing with data.
List of speakers who are nonmembers: None