http://www.wired.com/2015/06/answer-150-year-old-math-conundrum-brings-mystery/

I just came across this: it’s an interesting story about the history of Design Theory (which is connected to experimental design). Does anyone happen to know if any algebraic statisticians are working on Design Theory?

### Like this:

Like Loading...

JerzyOh, nice! On the applied side, I would guess that all possible such designs have been enumerated for small n, and that it’s rare to have an experiment so large that you need an algebraist to help you with the design.

But I bet people are still working on the theory side. Maybe here? http://library.iit.edu/jalgstat/

JerzyPS — we should get the department to run a 700-level experimental design course sometime. There is plenty of interesting stuff there, and it’s been unsexy for so long that we could probably write a ton of NIPS papers “inventing” it in an ML context 😛

kimolasPost authorUnfortunately I wasn’t able to find anything about Design Theory either on Google or through the J. Alg Stat, but yeah, it seems from the article that experimental design software already does something to find available (exact?) designs, so maybe it’s just a pure math thing from here on out. It seems as though for t > 2 there is still little known, but I’m not sure how relevant it is to experimental design.

I also think we should have an experimental design course for PhDs. I TAed the sampling and survey design course (300-level) last semester and I didn’t learn much besides how to write better surveys, unfortunately. I think there’s a ton of research being done in experimental design (some of which is by a subset of algebraic statisticians) and at CMU we haven’t been exposed to any of it.