Ihaka Lecture Series: Statistical computing in a (more) static environment
- Date: Wednesday 29 March 2017, 06:30PM to 29 March 2017, 07:30PM
- Location: Room DA02, Dovedale Village, Dovedale Campus
- Presented By: Ross Ihaka, Department of Statistics, University of Auckland
- Ticket: Free
Ten years ago, Duncan Temple Lang (perhaps jokingly) accused the creators of R of being guilty of “killing research in statistical computing”. Although appearing exaggerated at the time, the statement has turned out to be prophetic. Much of what passes for statistical computing research these days consists of little more than writing ‘R packages’. While there may be some merit in this kind of activity, research at a more fundamental level is desperately needed. This talk is concerned with that deeper level.
Statistical computing systems exist on a spectrum. From the very dynamic at one end, to the very static at the other. Dynamic systems (like R, S or LispStat) provide a very flexible computing environment but the price for this flexibility is performance. A lot is known about the dynamic type of system but only limited experience has been gained at the static end of the spectrum. In particular, not much is known about the flexibility/performance trade-off and whether more static systems can provide sufficient flexibility to justify their (potentially) improved performance. This talk will look at some of the problems associated with implementing a more static statistical computing environment.
Ross Ihaka is an Associate Professor of Statistics at the University of Auckland. Ross and Robert Gentleman were the originators of the R statistical software system. Ross’s research interests include statistical computing research and exposition, and also statistical graphics and visualisation.
This event is part of the Ihaka Lecture Series and is live streamed from University of Auckland’s Department of Statistics.
24 March 2017 - Event
Josh SchipperPhD CandidateElectric Power Engineering Centre (EPECentre)Department of Electrical and Computer Engineering, University of CanterburySupp...
24 May 2017 - Event
By their very nature, adventure sports involve managing risk, and rock climbing is an archetypal adventure sport. On leaving the ground, a climber has...