Roman M’s anomaly combination incorporated into R
Posted by Jeff Id on July 23, 2011
As long time readers know, I’m a fan of Roman’s temperature combination method which doesn’t require a base period window to offset individual station anomalies in global temperature averaging. Steve Mosher has incorporated Roman’s work into his own code for R.
steven mosher said
Nick Stokes code is up next! send rockstar and twinkies
Kenneth Fritsch said
I recommend following the link to Roman’s blog and reading his easily understood explanation of what his algorithm is attempting to estimate and the assumptions it entails. Stokes” and Tamino’s algorithms are, I believe, attempting to make the same estimates.
slimething said
Hypothesis:
All metrological aspects of [temperature] measurement have been considered and implemented prior to any further statistical analysis is considered.
At least that is what I was taught and could only report if true.
MIkeN said
Have you changed your hockey stick posts? I remembered it as some of the make a stick matches to sine waves and such had more proxies correlate than 400. Now I am seeing much less matches, and a higher correlation level.
This came up at Bart Verheggen’s when Jim Bouldin defended Mann’s paper and ridiculed your analysis.
Jeff Id said
Mike,
I’ve changed nothing and wouldn’t mind reading Jim’s criticism. Any links to where it happened?
MIkeN said
The same thread where you had some posts at the beginning. It took a turn through Tiljander and Mann 08.
Jeff Id said
I guess I’ll let Jim come here and tell me why it is wrong. I don’t have time now anyway.
Carrick said
MikeN, it’s a lot easier to make assertions (and to ridicule other people than deal with substance) and get away with it on a non-technical blog like Bart’s, than it is on an unmoderated technical one. When somebody engages in ridicule, my first assumption is they’re just stupid. We’ll see.