"Even after half a century of quantum theory and under the present impact of information theory and cybernetics, one often finds the rather outdated opinion that the application of statistical methods to a problem is dictated by our inability to solve it 'exactly'. Yet unless statistics are understood as a mere averaging process, the opposite is in fact true. A random quantity can be considered as found when not only its mean value, but the entire probability distribution about this mean value is known. In certain rare and usually idealized cases, this probability distribution is such that the mean value is, in the limit, the only value the quantity may assume and in this case the quantity is no longer random. In this special case of a single value instead of an entire distribution, we have a (so-called) 'exact' solution; this is of course much easier to find than the statistical solution, which includes the 'exact' solution as a special case. From this point of view the two opposites are not 'statistical' and 'exact solution', but 'general, statistical solution' and 'idealized limiting case'. "
YEAH! :)
Just wanted to write this out, because I think its particularly relevant today to the interfacing problem of science and society, and also because I keep wanting to remember it, and want to stop carrying the book around with me :)
"The Scattering of Electromagnetic Waves from Rough Surfaces" - Petr Beckmann, Andre Spizzichino, page 71.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment