I haven't followed your discussion too closely on the crosscorr
calculation, but I ran into a optimization bottleneck (perhaps similar
to yours) using an indicator to calculate the slope of the price
(least square fit).  The problem was that the indicator had to cycle
through all previous datapoints at each update.  My solution was to
subsample at a regular interval (skip parameter) while still using a
sliding window to make sure each point was considered.  So for a 20min
window (1200 points), I would subsample at 50 or 100 and still get a
nice estimate in a reasonable amount of time.  Perhaps there is a
better way, but I couldn't come up with it.

Also, maybe others would find this indicator useful, so I've uploaded
it:

http://groups.google.com/group/jbooktrader/web/RegressionSlopeEfficient.java

On Aug 5, 9:48 am, Eugene Kononov <[email protected]> wrote:
> Thanks, Shaggs, your LinkedList solution does work. It's still slower than
> the other "normal" indicators, but it's manageable.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"JBookTrader" 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/jbooktrader?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to