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: Wharf Rat who wrote (870504)	7/5/2015 12:15:56 AM
From: i-node	1 Recommendation Recommended By FJB Read Replies (1) \| Respond to of 1572631

fact....the hockey stick is real fact...Mann's papers were only crap in the denilosphere You either don't know what you're talking about or you are just refusing to admit the truth. I'd like to say Mann's papers were statistical nightmares, and they were -- but worse, they were data nightmares. It is important in inferential statistics that you not violate the basic rules with the data. That's before you even begin applying statistical methods to it. If you are missing data points, you're missing. You can't just infill them arbitrarily without adversely affecting the entire analysis. There were, literally, 100s of problems with these papers. Some minor, some, not so much. The bottom line, as I'm sure you're aware, is that Mann's model was such that whatever data you put into it, you'd get a hockey stick. Pretty much anything. This was a result of a clearly identified bug in his PCA centering algorithm. We know this; McIntyre proved it, very publicly. AFAIK, Mann NEVER made the requested information available to McIntyre who had to to dig it out piece by piece. The years of ongoing bullshit with the so called peer reviewed journals was absolutely outrageous. If you aren't familiar with it you should be as it totally makes it impossible to have any confidence in them. Most important, though, are the statistical failures with respect to temperature proxies. The hockey stick used proxies that should not have been used, proxies that are not even known to BE proxies, and as previously pointed out, proxies for which the entire dataset for some years consisted of a SINGLE tree. The so-called climate scientists are interested in getting the result they want, not the truth. The mathematics are simple enough for you to understand. I can only assume you are choosing not to find the truth.

To: Wharf Rat who wrote (870504)	7/5/2015 12:37:06 AM
From: i-node	1 Recommendation Recommended By FJB Read Replies (2) \| Respond to of 1572631

>> I'll bet that's why all studies use more than one tree FYI, the Gaspé from 1404 through 1421 was based on a SINGLE TREE. Only two trees were in that set from '21 thorugh '47. There is pretty good likelihood that is why the original study omitted those years. Mann comes along, and by using his buggy centering software, was able to control inclusion of the Medieval Warm Period. So, it was awfully convenient for him to use essentially a 47 years period based on one tree for a period, and two trees for the subsequent period -- by any account, statistically meaningless. Yet, it allowed him to control the content of the hockey stick formation because his software, having found any series with a hockey stick formation, was guaranteed to produce a hockey stock. What is worse is that the so-called peer reviewed journals had this presented to them time and again, concisely, and with convincing proof, and either didn't understand it or didn't care. In some cases (particularly, some of the citations you provided) they simply refused to listen and would publish papers they viewed favorably without a normal peer review process at all. All the while, Mann hiding out in his office, preventing access to his data and methodology. I would also point out that Mann used the so-called "RE" verification statistic, knowing full well he had an R^2 approaching zero, reflecting no meaningful correlation at all. Yet he moved forward, and he hid it. And the journals backed him. Even when his associate, Ammann, refused to release his R^2, it was eventually shown to be near zero. I'm telling you, there are a lot of problems with the statistics on this stuff, and these guys are going after outcomes and making it happen. I'm a skeptic, not a denier. But this nonsense, over time, makes me more doubtful. If they can't prove it without this making shit up -- and they haven't so far -- then you have to question whether it exists at all.