Citat:
Ursprungligen postat av
casefold
Jag lade till en edit i mitt förra inlägg.
Men PCAn bygger ju på val av proxys. Och vilka perioder som används för normalisering.
Their argument since the beginning has essentially not been about methodological issues at all, but about ‘source data’ issues. Particular concerns with the “bristlecone pine” data were addressed in the followup paper MBH99 but the fact remains that including these data improves the statistical validation over the 19th Century period and they therefore should be included.
If you use the MM05 convention and include all the significant PCs, you get the same answer. If you don’t use any PCA at all, you get the same answer. If you use a completely different methodology (i.e. Rutherford et al, 2005), you get basically the same answer. Only if you remove significant portions of the data do you get a different (and worse) answer.
https://www.realclimate.org/index.ph...k-controversy/
Citat:
Stephen McIntyre and Ross McKitrick (neither of whom are climate scientists) published a paper in which they claimed to invalidate the hockey stick graph on the grounds that it was supported by bad data (McIntyre and McKitrick 2003). They reached this conclusion by using an incorrect version of the proxy data set used in MBH98 (Rutherford et al. 2005, p. 2,312). In a subsequent article, they argued that the hockey stick is an artifact of the statistical conventions chosen in MBH98 (McIntyre and McKitrick 2005). Their argument in this paper targeted MBH’s use of principle component analysis (PCA), which is a statistical procedure that can be used to represent large data sets in terms of a smaller number of patterns. MBH used this procedure to deal with the problem of diverse proxy data, which included large sets of tree ring data and smaller sets of data from other sources (MBH98; Mann 2012). McIntyre and McKitrick (2005) argued that the hockey stick pattern is an artifact of the choice of a convention for centering tree ring data; more specifically, they argued that if one centers tree ring data at the twentieth century, the hockey stick pattern appears, but that if one centers it around a long term average (from 1,400–1,980), then the pattern does not appear. Like their argument that the hockey stick pattern is an artifact of bad data, this argument was deeply flawed and quickly exposed as such. MBH analyzed the data into principle components (PCs), one that describes periodic, oscillating temperature changes, and another that describes global warming. McIntyre and McKitrick adopted a retention criterion for PCs that had the effect of tossing out a global warming PC; “they had chosen to throw out a critical pattern in the data as if it were noise, when an objective analysis unambiguously identified it as a significant pattern” (Mann 2012, p. 138). This problem is confirmed in Wahl and Ammann (2007).
https://link.springer.com/article/10...194-014-0101-x
Men, återigen. Detta är off topic. Det har nämligen inte med förtalsrättegången att göra.
