Am 18.03.2012 12:02, schrieb Sergiu Ivanov:
I'm a bit sceptical about trying to get correlations out of user input.
Repeating myself: User input is multi-channel high-noise input, extracting
useful signals from that is hard.
Does your scepticism mean that you don't expect much useful
information without considerable effort or that you don't expect
anything useful of collecting the data at all?
It's the former: the information is in the data, but it's hard to
extract it.
Also, the statistics involved is nontrivial. You'd need a working body
of statistical expertise to interpret the correlations correctly.
There's the simple case of applying the rules incorrectly, and the
somewhat more advanced case that if you have 100 correlations, each with
a significance of 99%, you can expect one of them to be bogus (and if
you're unlucky, several are, and if you made some error, all of them are
except those that happened to be real).
--
You received this message because you are subscribed to the Google Groups
"sympy" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sympy?hl=en.
