Tuesday, 10 July 2012

Epistemological limits

New Scientist
5 July 2012

The confidence level of 7.4 sigma represents only random error and says nothing about systematic error, so I would not label it a "certainty." The importance of this distinction was dramatically demonstrated in the recent faster-than-light neutrino controversy. For this reason, I feel that 7 sigma versus 5 sigma is of little conceptual difference.

No comments:

Post a Comment