Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: J_F_Shepard who wrote (870979)	7/7/2015 1:41:47 PM
From: i-node	2 Recommendations Recommended By Brumar89 gamesmistress Read Replies (1) \| Respond to of 1578011

>> The statistics are real data..........interpreted by real scientists who choose not just one set but study thousands of sets...... Not thousands. The Mann paper in '98, that gave rise to the "Hockey Stick" misnomer, was based on 116 data series. From those, Principal Component Analysis (PCA) was used to narrow it down to "several", and of those "several" only a few had meaningful levels of correlation. But the important point is this: Mann, et al., made a mistake by using what is referred to as "short centering" for critical datasets. This process was discovered by others examining his FORTRAN source code. The interesting thing is this: When McIntyre & McKitrick applied Mann's technique of short centering to random data ("Red noise") they found that it also produced a hockey stick. And in fact, if any single series of data included in the analysis produced a hockey stick, then it would always generate a hockey stick for the entire analysis. Given the cherrypicking, which was admitted before a congressional subcommittee, with the claim it was "appropriate" ("If you are going to make a cherry pie, you're going to pick cherries") by members of the Hockey Team, Mann could always be assured of a hockey stick and if he didn't get one, he would just change the included datasets to guarantee it. After all, the correlation figures were rigged, yielding R^2 of close to zero for many of the series on which he relied heavily. It is bullshit. That's not to say that someday, convincing proof couldn't turn up. But as it now stands, this amounts to a cabal of people with PhDs who are manipulating data to get a desired result. >> But I am amused at your declaration that Walmart is a great company in the context of big oil, high tech, computers, autos, aircraft, military, etc........ Walmart is nothing but a large friggin general store.......... The world needs general stores, too. Walmart is, by any measure, a "great company". That it isn't in a technology field is neither here nor there. Not all great companies are tech companies. Kind of a stupid proposition, if you ask me.