| Revision as of 21:14, 25 October 2004 editR. fiend (talk | contribs)Extended confirmed users, Rollbackers24,209 editsNo edit summary← Previous edit | Revision as of 12:08, 2 November 2004 edit undoWilliam M. Connolley (talk | contribs)Autopatrolled, Extended confirmed users, Pending changes reviewers, Rollbackers66,050 edits Insert text from T rec of last 100 y verbatim, before editing itNext edit → | ||
| Line 4: | Line 4: | ||
| McKitrick also wrote , coauthored with Christopher Essex and published by Key Porter Books. It was runner-up for the Donner Prize as the Best Canadian Book on Public Policy and finalist for the Canadian Science Writers' Association Book Prize. | McKitrick also wrote , coauthored with Christopher Essex and published by Key Porter Books. It was runner-up for the Donner Prize as the Best Canadian Book on Public Policy and finalist for the Canadian Science Writers' Association Book Prize. | ||
| === Monte Carlo Counterproof === | |||
| There is a proof from the canadians ] und ], that indicates fundamental computation errors. Surprisingly the publication of those ] based insights were rejected by the well known magazine ]. The Physics ] meanwhile was able to publish in ] the proof of ] und ] that nearly any data, e.g. data that is totally random, set fed into the computation program did lead to the hockey stick curve. This method is well known in the subject of computer science by the name of ], which just means to randomize data in a simliar way as in the Casion of ]. His research showed that there was a library in use which was only valid vor dates of the year 1902 and later. Handling for all former data was not specified for that program which means that the module would produce results of a different nature than expected. Other than that there were a few more problems unveiled when the program got analyzed, which in short were able to produce the supposed trend, even when there is no trend, by just concentrating more on data that fits compared to data that does not fit. (Compareable systematic are well known from situations in medical sectors where it was told to data eximinars that one group got a placebo whilst another got the good substance, wilst in fact both groups were using the placebo, but the created report showed a clear benefit for the substance.) | |||
| Muller explicitely states that this error is of such a fundamental category that the current results are rendered useless and the whole computation must be redone for achiving overall credibilty once again. Despite this he does not see a reason for doubts on the general agreement on the effect of global warming, even if this single result has to be pulled back. | |||
| == External links == | == External links == | ||
Revision as of 12:08, 2 November 2004
Ross McKitrick is an economist who has recently (since approximately 2002) worked on global warming; in this connection he is a skeptic. His best-known work is Corrections to the Mann et. al. (1998) Proxy Data Base and Northern Hemispheric Average Temperature Series, coauthored with Stephen McIntyre. This presented an "audit" of work by Michael Mann and others; it appeared in the minor journal "Energy and Environment" in 2003. Mann et al. have replied, doubting the validity of McIntyre and McKitrick's work . The matter was later referred to Nature, following which Mann et al. published a corrigendum, including a re-statement of their data and methods, which appeared on July 1, 2004. The corrigendum did not affect the results, however. McKitrick et al. wrote up a longer version of their results, which was rejected by Nature.
McKitrick's own data analysis has been strongly criticised, in particular by Tim Lambert, who has written on his web page about what he considers serious flaws in some of McKitrick's publications. Most notable is a failure to distinguish radians and degrees which Lambert claimed invalidates the conclusions of one of McKitricks papers. The author's response (nb this is the authors response; it is not yet clear whether the journal will accept it) acknowledges the programming error (it glides over the mistaking of degrees and radians as "There was a small error" without acknowleding the details) but asserts the effects were minor and did not undermine the paper's conclusions.
McKitrick also wrote Taken By Storm: The Troubled Science, Policy and Politics of Global Warming, coauthored with Christopher Essex and published by Key Porter Books. It was runner-up for the Donner Prize as the Best Canadian Book on Public Policy and finalist for the Canadian Science Writers' Association Book Prize.
Monte Carlo Counterproof
There is a proof from the canadians Stephen McIntyre und Ross McKitrick, that indicates fundamental computation errors. Surprisingly the publication of those computer science based insights were rejected by the well known magazine Nature. The Physics Richard A. Muller meanwhile was able to publish in 2004 the proof of McIntyre und McKitrick that nearly any data, e.g. data that is totally random, set fed into the computation program did lead to the hockey stick curve. This method is well known in the subject of computer science by the name of Monte Carlo method, which just means to randomize data in a simliar way as in the Casion of Monte Carlo. His research showed that there was a library in use which was only valid vor dates of the year 1902 and later. Handling for all former data was not specified for that program which means that the module would produce results of a different nature than expected. Other than that there were a few more problems unveiled when the program got analyzed, which in short were able to produce the supposed trend, even when there is no trend, by just concentrating more on data that fits compared to data that does not fit. (Compareable systematic are well known from situations in medical sectors where it was told to data eximinars that one group got a placebo whilst another got the good substance, wilst in fact both groups were using the placebo, but the created report showed a clear benefit for the substance.)
Muller explicitely states that this error is of such a fundamental category that the current results are rendered useless and the whole computation must be redone for achiving overall credibilty once again. Despite this he does not see a reason for doubts on the general agreement on the effect of global warming, even if this single result has to be pulled back.
External links
- http://www.uoguelph.ca/~rmckitri/ross.html - his home page
- http://www.takenbystorm.info/ - home page for his book, "Taken by storm"
- http://www.uoguelph.ca/~rmckitri/research/trc.html - "Corrections to the Mann et al (1998) Proxy Data Base and Northern Hemisphere Average Temperature Series"
- http://www.climate2003.com/ - home page of his coauthor Stephen McIntyre
This article is a stub. You can help Misplaced Pages by expanding it. |