I always use brute force. D/C is a lie, in my opinion. I start with a wide and coarse pass, look for a "good area", and then shrink the region and increase the density.
If your good regions have a lot of variance, you probably don't have enough trades to make the optimization peak statistically that meaningful. it is likely your number of trades is low and the slight change of the parameter cherry picked a couple trades in ways that you can't extrapolate to the future. Your better off just shooting for the center of the good region than looking for a peak, in my opinion, and avoiding danger zones that might "migrate towards you" in the future. If the number of trades is low (which is dangerous in my opinion), it's very informative to grab a few points in the region with different results, do a backtest on them, and compare the graphs (you may need to do a screenshot for this). Lots of interesting information can come out of this - such as the cherry picking I was describing - where certain trades show up in some parameter sets, but not in others. This can also make you go mad though :P On Sat, Aug 15, 2009 at 9:30 AM, JBTrader <[email protected]> wrote: > > I cought myself that I almost always use D&C.... wondering if you guys > are using it?.... > > > --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
