Comments on: Juckes and the Esper and Jones CVMs

By: Willis on “Getting authors to respond to questions” « Climate Audit

Sat, 04 Dec 2010 18:14:02 +0000

[...] (uckes and the Esper and Jones CVMs) [...]

By: UC

UC — Fri, 03 Nov 2006 17:48:17 +0000

#19

Yes, good point, check also http://www.climateaudit.org/?p=886, #19, #26. The latter is also a good warning for those who use calibration residuals for uncertainty range estimation.

To me it seems that inverse regression (global temp) tends to overfit, and CVM cannot handle sampling problems. Local calibration works well, and it gives directly estimate of proxy noise.

By: Phil B.

Phil B. — Fri, 03 Nov 2006 16:53:24 +0000

Re #18, UC, small point but normally I’d use matlab’s cmd pinv(Bh) instead of inv(Bh’*Bh)*Bh’, especially if Bh might ever be poorly conditioned.

By: UC

UC — Thu, 02 Nov 2006 18:23:44 +0000

OK, thanks. And Inverse Regression, as described in A1, in Matlab format would be:

yc=y(1:NCal)-mean(y(1:NCal)); % calibration data, global temperature (?)

Xc=X(1:NCal,:); % X is a matrix of standardized proxy records, calibration period 1:NCal
Bh=(inv(yc’*yc)*yc’*Xc)’; % estimate of beta
yh=inv(Bh’*Bh)*Bh’*X’; % reconstruction

I think it is not too hard to write m-file that simulates different scenariors.. Let’s see..

By: Steve McIntyre

Steve McIntyre — Thu, 02 Nov 2006 17:40:55 +0000

Yes. I’ll post up a script showing my replication of the Euro Team composites. In this case, it looks like they’ve scaled on 1856-1980 – ao that there is no “verification period”, but I might have not got it exactly.

By: UC

UC — Thu, 02 Nov 2006 17:37:16 +0000

Steve,

did I understand CVM correctly:

1) average the standardized proxies

2) standardize the average, and scale so that calibration data mean and variance and reconstruction data mean and variance (within calibration interval) match?

By: jae

jae — Thu, 02 Nov 2006 16:34:34 +0000

I guess I’m dense. These studies simply do not follow the scientific method and should be rejected on that basis alone. I just don’t understand how any scientist can justify selecting only certain series that conform to their “expectations,” without some documented a priori basis for doing so. It must be that the tree ringers have somehow tacitly accepted some really bad scientific procedures. Maybe they actually think they are justified in doing this, because it is a standard procedure in dendrochronology.

By: fFreddy

fFreddy — Thu, 02 Nov 2006 16:30:12 +0000

(These RCS calculations are mathematically trivial, but the ringers make them out to be some kind of magic.)

Which tells you something about their preferred audience.

By: bender

bender — Thu, 02 Nov 2006 16:03:06 +0000

On publication. Auditing is impossible without adequate documentation. That’s why they invented shredders. I’m amazed you’ve gotten as far as you have.

On RCS. Some people have to resort to making things sound like magic in order to get them on the shelf of the marketplace of ideas. The ones who just quietly go about their science don’t get the big grants.

By: Steve McIntyre

Steve McIntyre — Thu, 02 Nov 2006 15:50:25 +0000

#11. At least the Euro Team have provide enough documentation that one can more or less see what they’re doing and at least attempt sensitivity studies of this type. It’s all very well for TCO to argue that I should have done this for the other studies, but it’s not as easy as all that when I didn’t even have the Esper chronologies, which underpin other results. It also makes a big difference having Esper’s version of the Polar Urals Update. I can substantially replicate his calculation, but you can imagine the caterwauling that would have taken place had I presented a sensitivity study with my RCS calculation of Polar URals update. (These RCS calculations are mathematically trivial, but the ringers make them out to be some kind of magic.)