View Single Post
Old 12-02-2009, 07:20 PM   #209
XNN
Junior Member
XNN has a complete set of Star Wars action figures.XNN has a complete set of Star Wars action figures.XNN has a complete set of Star Wars action figures.XNN has a complete set of Star Wars action figures.XNN has a complete set of Star Wars action figures.
 
Posts: 7
Karma: 400
Join Date: Dec 2009
Device: none
Quote:
Originally Posted by Dulin's Books View Post
XNN Have you read the NAS and Wegman reports about Mann's hockey stick paper?
Sorry, no I haven't. Paleo work is well outside my direct area of expertise.

Quote:
Please since you are in statistics I'd like your opinion of the method used for Wegman and also if you could your remarks about rasmus response to post 19 here
Linked quote and reply:

Quote:
From link: "Re #10 response: Gavin, you forgot to mention the other sort-of original part of the Wegman recommendations, which is to employ a lot more statisticians in climate science research.

[Response:I have yet to see any evidence for the implicit assertion that statistitians bring forth the true answers whereas other scientists don't. My guess is that there are clever statisticians as well as poor statisticians - probably as in other disciplines. There are as far as I know a number of statisticians involved in climate research already, and they bring in benefitial new aspects to the analysis, but physicists are just as important (if not more) for advancing the climate science. I sounds as if Wegman has some naive idea are some kind of magicians who can bring out the right numbers (?). -rasmus]"
I'm of two minds here. I have seen truly disastrous -- at least from the standpoint of statistical methods -- papers by pure climatologists get past peer review. These definitely would have benefited from a statistician's eye. On the other hand, I have seen pure statisticians try to cross over into applied climatology and make some truly boneheaded conclusions because they didn't have a basic grounding in the physical sciences.

The strongest work tends to be from researchers with a solid background in both disciplines, or by collaborations between the two camps. In Canada (where I work) the climate research community has, possibly by chance, fallen into this mold. Given the size of the community, I think that this has allowed us to punch above our weight class, so to speak. As an example, the former head of the national climate modeling center and current director of climate research within the government is a statistician.
XNN is offline