Monday, June 25, 2012

BayesComp2012

David Benson asked what we were doing in the company of mathematicians at Tachikawa...

There's a huge Bayesian stats meeting in Kyoto this week, which would probably have been quite interesting but which I thought was a little too far away both geographically, and topically, to be worth attending, especially right now as I'm pretty busy.

Fortunately, a satellite meeting was arranged last Friday/Saturday, hosted by the Institute of Statistical Mathematics in Tachikawa (which I once visited at its previous site in Hiroo in central Tokyo, before they moved). Some people there are also working on climate change related projects, possibly the same projects we are working on though there seems to be some overlap/duplication between the various ministries and institutes as to who is doing what! Several of the eminent attendees of the Kyoto meeting were somehow persuaded to come a few days early and visit the fleshpots of Tachikawa - I hope they thought it was worthwhile - but for us it was a great opportunity to hear what is going on in the latest research into computational methods for the Bayesian paradigm (mostly Markov Chain and Sequential Monte Carlo methods). And it was also a good excuse to visit the new Tachikawa site of ISM, and realise that it's one more place where we really wouldn't much want to work when our time here in Yokohama is up :-)

The meeting was, as expected, a little obscure and distant from our work - which confirmed my decision to not go to Kyoto - but was well worth spending a couple of days on. One or two bits were particularly interesting - especially the methods for estimating very small probabilities (down to 10-120 or even 10-200), which may be relevant to our future plans, now we are in a post-Fukushima world and being urged to plan for the unimaginable...

12 comments:

guthrie said...

I had a look at Curry's website today and found her touting a paper of Mkitrick's:
https://springerlink3.metapress.com/content/4331350766569165/resource-secured/?target=fulltext.pdf&sid=uesqfzyso4iydpkvotfu2a4x&sh=www.springerlink.com

Apparently they do some Bayesian analysis. Maybe that is up your street?

Steve Bloom said...

I prefer imagining the unplannable. :)

Guthrie, Gavin already trashed that paper, albeit not at length, over at RC (in the June open thread IIRC). The basic problem seems to be that it continues the fundamental error of McK's past similar efforts, Bayesian lipstick notwithstanding.

That leaves us to wonder what the editors and peer reviewers were thinking. Fair play in science is a wonderful thing, but at some point these people have acquired enough form to justify telling them no the moment they heave into view. The PNAS response to Lindzen's latest effort shows the way.

guthrie said...

Ah thanks, I hadn't read all of that thread.
Yes, you'd think that their past history would result in their paper getting very close attention.

James Annan said...

Thanks, I will check the comments. I'd seen the paper and recognised the history, but not actually read it yet. It doesn't look from the abstract that they are actually making very strong claims, though, just a non-zero socioeconomic effect (which may be v small and/or aliased from something else).

James Annan said...

Oh good grief. That didn't take long. Gavin is correct that using ensemble averages (of arbitrary size) in the way they did makes the whole thing a waste of time.

crandles said...

Planning for the literally unimaginable must be difficult :o)

I don't see how you can calculate a probability, even if you can go down to 10E-120, for something unimaginable.

Imagining the unplanable :) yep that is fun. You can imagine all sort of chaos but even if things are equally likely to go in opposite directions you can still plan for both directions and also plan to try to make society more resilient to change. So you have to be imaginative :)

More seriously, what do they mean by unimaginable?

Where does something like Arctic becoming seasonally ice free in the next few years fit in. Probable enough to get some serious planning? Can GCMs be forced with sea ice extent and snow lines with any hope of getting some idea of climate changes that might result?

James Annan said...

Being Japan, a new hyper-typhoon sitting on top of Tokyo is the sort of thing they are thinking of. But once you abandon any concept of probability, it becomes a bit ridiculous - just a game of who can invent the greatest catastrophe. In that context, a meteor or new disease would rank significantly higher than climate change IMO.

Japan already is pretty resilient to extreme climate - 500mm of rain in a day is not unheard of, and although it causes some localised problems, it's generally not a major catastrophe.

David B. Benson said...

Ah, consorting with statiticians, not real mathematicians.

:-)

jules said...

>Ah, consorting with statiticians, not real mathematicians.

Yes I was wondering about that. Are the statisticians kind of the dregs of mathematics or something? The whole no lunch on the second day and posters that it was phyiscally impossibe to get close enough to read was quiet disconcerting. THat level of disconnection with reality makes climate scientists seem positively human by comparison.

James Annan said...

I suspect it's not atypical for mathematicians generally.

David B. Benson said...

Mathematicians always do lunch and AFAIK never do posters.

crandles said...

>"Japan already is pretty resilient to extreme climate - 500mm of rain in a day is not unheard of, and although it causes some localised problems, it's generally not a major catastrophe."

So the more likely catastrophe is that the storm tracks move so that Japan avoids serious weather events but other countries that are not used to such events get them?

Meanwhile the planning goes into Japan getting worse weather.