Also, maybe it would be worth weighting the "importance" of each day. For example, P/L values 3 weeks ago should have less importance than P/L yesterday...
I'm just throwing ideas to come up with more "forward looking" parameters for the strategy thoughts? On Aug 4, 11:20 pm, JBTrader <[email protected]> wrote: > Talking of optimization... have you considered something lke walk > forward optiization... > for example, we specify number of past days to use for optyimization > and number of future days to test > click the button and sit back. > > So after couple of iterations we determine optimal days required to > fit the strategy > in order to maximize short term "future" performance? > > On Aug 4, 11:17 pm, JBTrader <[email protected]> wrote: > > > Thanks for explaining that Eugene > > > On Aug 4, 9:50 pm, Eugene Kononov <[email protected]> wrote: > > > > > by exponentialy I meant that initialy it estimates 35minutes (for > > > > example) and in 10 minutes it says 15 rather than 25minutes remaining. > > > > > its not exacly "exponentialy" but it decreases faster than "real time" > > > > goes... if you know what I mean... > > > > Oh, ok, I thought you were referring to D&C accuracy with respect to > > > finding > > > the top parameters. > > > > Yes, the "time to completion" estimate in D&C is not accurate. That's > > > because with D&C, it's impossible to know in advance how many strategies > > > it > > > would try before the search space converges into a single point. --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
