lovely reception, slightly less lovely weather

The RSS conference social programme got underway this evening with a lovely reception at the Sea Life Centre in Brighton. The centre is a large aquarium with a series of interesting displays of aquatic life, and the venue was open this evening only for participants of the conference. So we had the run of the place, enjoying juice, wine, and hors d'oeuvres, while marveling at the undersea world and reveling in the joy of renewing old acquaintances and making new ones.

Registration began this afternoon, and in typical sharp RSS fashion, it was well organized. I was in and out of registration in under a minute, but spent several very pleasant minutes afterward meeting and greeting RSS staff.

The weather got quite blustery and then rather wet this afternoon. Medium-sized waves crashed loudly against the beach and the Brighton Pier. It is a bit chilly as well, or so it seems to someone coming from the lingering summer weather in the US. Nonetheless, the coast and the town are spectacular, a great setting for the conference.

The scientific programme begins tomorrow with opening remarks from David Hand, RSS President and a plenary talk by Peter Donnelly of Oxford. We’ll all be there Brighton-early (sorry, I have a rather nasty addiction to puns).

Statistical Engineering and a theory that dates back to 1914

One of the best methods that I have come across which exemplifies the inductive-deductive iterative nature of statistical investigations (see my first post) dates back to 1914 – the so-called “Pi” theorem of E Buckingham; I will illustrate the use of the “Pi” theorem using the well known paper-helicopter experiment, which many people who have taught statistical methods to engineers will be familiar with. If we adopt a completely empirical approach, we might decide to run a response surface experiment to model the flight time of the helicopter as a function of various design parameters; three design parameters might require about 15 runs in the experiment to develop the transfer function. However, if we think for a minute about the physics, we know that the flight time will be a function of the mass of the helicopter, and the area swept out by the rotors, together with the force due to gravity, and the density of air – and all of these quantities are known. The application of the “Pi” theorem, which reduces the dimensionality of the problem, and does not require linearity to ensure dimensional consistency, reveals that the number of experimental runs can be reduced to about three. It is a mystery as to why the “Pi” theorem isn’t referenced in any of the classic texts on response surface methodology and design of experiments; is it because not enough statisticians are interested in engineering?

getstats zone at RSS 2010

The getstats team are looking forward to RSS 2010 in Brighton next week. All of the delegates are encouraged to visit the getstats zone which will have information about the campaign and the ways you can get involved and answer any questions. We are also going to be doing some filming of vox pops to get everyone’s views on the importance of statistics and the campaign. Please drop by the zone to find out more! We look forward to seeing you there.

Robustness in Engineering

In engineering, reliability problems come about for essentially only two reasons 1) mistakes, and 2) lack of robustness. Genichi Taguchi did much to bring to our attention the idea of robustness (making designs insensitive to variation, or “noises”), although others had been there too, notably RSS Fellow and Greenfield medallist Jim Morrison as far back as 1957. Taguchi had some important things to say about strategies for improving robustness, one being that engineers should first look to desensitize their designs to variation through experimenting with design parameters related to geometry, material properties and the like, and not to choose the more obvious path of trying to reduce or eliminate the noises. I will explain some of Taguchi’s ideas, and hope to demonstrate that he didn’t deserve some of the attacks on him by the statistical profession at the time, in stark contrast to the way our profession seems to have embraced the Six Sigma movement with nothing like the same scrutiny afforded to Taguchi’s work.

Tim Davis

The role of likelihood in statistical science

There have been many interesting developments in theoretical statistics since you were in graduate school, but do these have any relevance for the practice of statistics? What is needed to make the translation from nice new mathematics to "on-the-ground"improvements? In my talk I survey some of the advances in likelihood-based inference, and try to identify the most promising links to better analysis of data.

Nancy Reid

University of Toronto

Predicting Credit Default Rates

Predicting spatial processes often involve using many, many parameters. That approach requires using Bayesian methods -- or something with the same effect -- to shrink the predictions back to something more reasonable. I'm going to use something simpler, regression. No, not a ridge estimator either. Rather, by constructing a particular explanatory variable, I can achieve much the same effect at the cost of just a few parameter estimates. My talk will cover this trick as well as show a variety of maps of the evolution of default rates in the US. Hope this is enough to lure you back inside from the beach next week!

Statistical Engineering & Reliability

High profile cases like BP, Toyota, and Firestone bring into sharp relief the subject of engineering for reliability. As statisticians, we seem to have got everybody from ourselves, to scientists & engineers, to senior management and to regulatory authorities, comfortable with the idea of expressing reliability as a probability. Indeed, in media interviews, the BP CEO quoted a failure probability of “about 10^-5” for the oil rig that exploded causing the spill. In his investigation into the 1986 Challenger disaster, when NASA management had quoted a similar probability for the reliability of the Space Shuttle, Richard Feynman said in his report into the accident “What is the cause of management's fantastic faith in the machinery?” Probability measures for reliability may be appropriate for some fields of engineering, but I will introduce an information based definition that is better suited to many engineering situations (including automotive) where the probability definition simply can’t be measured. I will argue that the focus should be on evaluating the efficacy of counter measures for identified potential failure modes, and the statistical methods required to evaluate this efficacy are much different to those required in attempting to measure reliability through a probability.

Tim Davis