|
PureBytes Links
Trading Reference Links
|
cross-correlation, in the investment
context, is typically used to measure the "closeness" of one function to
another and/or to identify leading and lagging indicators. autocorrelation
isn't terribly meaningful for most folks, again in context. I am not aware
of an "aggregating" property to correlation except in the sense it multiplies
the functions and sums the result.
There are certainly uses for the cross-correlation
function. it is fairly straightforward to implement in excel and its
counter parts. It is certainly useful for identifying colinearity between
instruments. The typical first hope of a time series fellow when he hits
stocks is to look for significant leading cross-correlation peaks between a
universe of stocks - looking for a Gillette/Schick or perhaps Bud/Miller
game. Lends its self to major number crunching types.
I suggest is may be fruitful to identify
economic/investment/chart quirks and characteristics and then find the toolto
exploit that. The other, trying all interesting operations, is likelyto
be at best a shot gun approach - and perhaps a random shotgun in the dark
<g>.
Cheers,
Richard
<BLOCKQUOTE
>
----- Original Message -----
<DIV
>From:
John
Nelson
To: <A title=amibroker@xxxxxxxxxx
href="">amibroker@xxxxxxxxxxxxxxx
Sent: Friday, July 12, 2002 7:52 PM
Subject: [amibroker] Corrolation
function?
All,Is there a Matlab-like corrolation function
in AFL (or available to AFL)? An auto-corrolator would be cool to
combined multiple signals to yield an aggregate signal (or so the thinking
goes). Anyone looked into this sort of thing?--
John--
_____________________________________________________John T.
NelsonTrader
| Dreams Of
Empiremail:
|
trader@xxxxweb:
| <A
href="">http://www.dreamsofempire.com/_____________________________________________________"Prediction
is sexy... but strategy makes money"Your
use of Yahoo! Groups is subject to the <A
href="">Yahoo! Terms of Service.
|