I would be curious on what responses, you get, since I have also witnessed some randomness for the same backtest parameters. What I have noticed is that when a run backtest a second time after the first (following optimization) the results are different.
--- In [email protected], "graphman27" <st...@...> wrote: > > Greetings: I've only had this happen a few times, but I run an optimization, > choose the best parameters, insert them as the new default values and run > another backtest. 98% of the time, with all my different strategies, I can > duplicate the optimized results, which is obviously what should happen. > However, on a few occasions, I follow those steps and get results not even > close to the optimized values. > > I'm assuming there is something wrong with the simple code I use in this > strategy. Maybe something with the "Buy" or "SELL" segment and not properly > using "CROSS." Being a newbie to AmiBroker, I can't see the problem, but I'm > sure someone with good afl skills will probably immediately spot it: > > EMAShortBuy=Optimize("EMAShortBuy",10,4,10,2); > EMALongBuy=Optimize("EMALongBuy",50,20,60,10); > StochRange1=Optimize("StochK1",20,10,40,10); > StochRange2=Optimize("StochK2",2,1,3,1); > > Buy = Cross(EMA( Close,EMAShortBuy ),EMA( Close,EMALongBuy )); > > Sell = StochK(StochRange1,StochRange2)>75; > > Short = 0; > > Cover = 0; > > Plot( EMA ( Close,10 ),"Short EMA", colorGreen, styleThick ); > Plot( EMA ( Close,50 ),"Long EMA", colorRed, styleThick ); > Plot( StochK ( 25,2 ),"stochk", colorBlue, styleThick ); > > /* max loss stop optimization */ > > ApplyStop(stopTypeLoss, > stopModePercent, > Optimize( "max. loss stop level", 0, 2, 10, 2 ), > True ); > > ApplyStop(stopTypeTrailing, > stopModeRisk, > Optimize( "trailing loss stop level", 0, 10, 90, 10 ), > True ); > > Any ideas are welcomed. >
