Thursday, November 18, 2010

Persistent Red Noise

Hokey Shticks, Unprincipled Proponents, and Spurious Significance

-- by Horatio Algeranon

Ed Wegman is a copycat
There's really no debating that

His famed report has plagiarized
His stats were McIntyre-devised

Though sold as “Independent Review”
There's really nothing there that's new.

In essence, it is just rehash
"Repete" with drama and panache.

Noise with perfect autocorrelation
To infinity without vacation.



Dedicated to Johnny Deep

Based on the findings of John Mashey and Deep Climate (here and here, for example) and inspired by a comment at Tamino's by Steve Metzler:
 
DC has proven that the ‘statistics’ were just a re-run of McIntyre’s code, using McIntyre’s saved data! So in essence, even the statistics were plagiarised :-)

..and also inspired by this:

"their method [that of Mann, Bradley, and Hughes] when tested on persistent red noise, nearly always produces a hockey stick shaped first principal component*..." -- McIntyre &McKitrick
*but M&M greatly exaggerated the effect by assuming climate "noise" with uncharacteristically (unrealistically) high/long "persistence" and by "illustrating" with "samples" drawn from a stored compilation of the most "hockey sticky" of the results (top 1%)  --as illuminated by Deep Climate (here and here)  and David Ritson here.

Horatio actually prefers the following statement (which, as it turns out, is about as meaningful as the M&M version)

"the method of Mann, Bradley, and Hughes when tested on persistent red nose, nearly always produces a jolly old elf (and his "team", of course)"
The latter is illustrated here. (Note the shape)


Update: Nov 23, '10

Experts claim 2006 climate report plagiarized (USA Today)

Climate science critic responds to allegations (USA Today)