SEARCH BLOG: POLITICS and GLOBAL WARMING
From my email... I don't necessarily subscribe to the political analysis and I believe the author does not subscribe to it as well. It was only done to make a point about bad statistical analysis. I would have further pointed out that the main deficiencies in the CO2 versus temperature correlation is that:
- CO2 increases after temperatures increase reversing the popular myth that CO2 causes temperature increases
- CO2 as a greenhouse gas is a minuscule portion of total greenhouse gases and even with the recent increases related to warming of 800 years ago, CO2 percentage of total atmospheric gases is basically unchanged
- Any greenhouse effect of CO2 is limited to a very narrow range and is self-limiting
from Steve Goddard date Thu, Mar 5, 2009 at 11:20 PM subject Blog for publication I got your name from Roger Pielke Sr., who suggested that I send this over to you. Normally I publish my articles on Anthony Watt's site, but this one is too political for him. Please let me know what you think.Thanks,Steven
- Steven GoddardConclusion Driven AnalysisThe news is littered with stories about scientists who have discovered damaging effects of global warming. Many of these analyses consisted of assuming global warming to be occurring, and then searching for data which appeared to fit the assumption.Let us make a similar analysis, where we assume something to be true and prove it by finding the right data. The assumption in this case being that President Obama is responsible for the stock market crash. As you can see below, the conclusion correlates at least as well with the data as many of the conclusions of peer-reviewed global warming papers.The graph above shows that if we ignore all other factors, there is excellent correlation between the highlights of President Obama's political success and the 50% loss of stock market value. Much better than the consensus correlation between CO2 and temperature, seen below.The IPCC claims 90% confidence between observed warming and CO2, and that is accepted by government - so I claim 99+% confidence in my analysis. The point being that if you start with a pre-conceived notion and consider only one degree of freedom in a multi-variate analysis, you can prove almost anything you want - whether or not it is true.For the record - analyst Obama gave his buy recommendation two days ago, based on a new terminology he coined "profit and earnings ratio."