Free Republic
Browse · Search
General/Chat
Topics · Post Article

To: shibumi
As they reported this Thursday, they have now reduced the uncertainty in their experiment to a level of 3.9 sigma - equivalent to a 0.005% probability that the effect is a fluke.

But particle physics has a strict definition for what may be called a discovery - the "five sigma" level of certainty, or about a 0.00003% chance that the effect is not real - which the team must show before they can claim to have solved the long-standing matter/antimatter mystery.

Could someone ask Al Gore what the sigma certainty of Anthropocentric Global warming is?

7 posted on 07/07/2011 2:30:03 AM PDT by ALPAPilot
[ Post Reply | Private Reply | To 1 | View Replies ]


To: ALPAPilot
Could someone ask Al Gore what the sigma certainty of Anthropocentric Global warming is?

You can ask anything you want, I suppose. In hypothesis testing, one must have a clearly stated set of hypotheses, at least two. In the case of matter, antimatter, I believe the two hypotheses are:

Null hypothesis: Matter, as opposed to antimatter, preferentially generated from the interaction of high energy electromagnetic radiation.

Alternate hypothesis: The null hypothesis is false.

One then gathers data, by conducting an experiement, or analyzing historical data, or both, to test the hypothesis and compares the results to the results expected if either hypothesis were true. For instance, if the Null hypothesis is that a given coin is "fair", exhibits heads and tails with equal likelihood, then the experiment would consisted in repeated trials (coin flips) to see if the results are consistent with a fair coin.

If we flipped a hypothetically fair coin ten times and it came up heads five times, we would not have "proven" that the coin is fair, only that our observations were consistent with a fair coin. If if came up heads (or tails) ten straight times, we would not have "proven" that the coin is unfair, only that the observations are highly unlikely (once in 1024 series of trials) given a fair coin.

As the number of coin flips became larger and larger, our certainty about the nature of the coin (see Bayes ninth proposition) would generally increase, though there would always be limits on our confidence. In a run of a thousand trials, a fair coin will produce between 469 and 531 heads 95% of the time. If our results fall in the range, we say with 95% confidence, that the coin is fair.

In global warming, the carnie pulls a coin out of his pocket, won't let anyone examine it, doesn't give you an opportunity to decline to play, tosses the coin and calls out "Heads I win, tails you lose!"

10 posted on 07/07/2011 3:13:55 AM PDT by Lonesome in Massachussets (Somewhere in Kenya a village is missing its idiot)
[ Post Reply | Private Reply | To 7 | View Replies ]

Free Republic
Browse · Search
General/Chat
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson