Машинное обучение в трейдинге: теория, модели, практика и алготорговля - страница 1147

[Maxim Dmitrievsky]

FxTrader562:

yes

[Maxim Dmitrievsky]

FxTrader562:

Actually, it would be really great if you can do some changes here...

which changes 

[Maxim Dmitrievsky]

FxTrader562:

Actually, it would be great if you could change here:

I tried this ... but I didn’t know if I did something wrong:

Please see the code and suggest if we can do something similar to improve the pip expectancy per trade

Better if you change here

//+------------------------------------------------------------------+
//|Get trade signal                                                  |
//+------------------------------------------------------------------+
double CRLAgent::getTradeSignal(double &featuresValues[])
  {
   if(!random_policy) {
      double kerfeatures[];
      ArrayResize(kerfeatures,ArrayRange(bestFeatures,0));
      ArrayInitialize(kerfeatures,0);
      
      for(int i=0;i<ArraySize(kerfeatures);i++) {
         kerfeatures[i] = featuresValues[0]/featuresValues[(int)bestFeatures[i][1]];
        }
            
      CLogit::MNLProcess(Lmodel,kerfeatures,Lout);
      return Lout[1];
     }
   else if(rand()/32767.0<0.5) return 0; else return 1;
  }

if(fast MA>slow MA) if(rand()/32767.0<0.8) return 0; else return 1;

if(fast MA<slow MA) if(rand()/32767.0<0.8) return 1; else return 0;

 or something like this, any indicator as filter

[FxTrader562]

Maxim Dmitrievsky :

Better if you change here

if (fast MA > slow MA)  if ( rand () / 32767.0 < 0.8 ) return 0 ; else return 1 ;

if (fast MA < slow MA)  if ( rand () / 32767.0 < 0.8 )  return 1 ; else return 0 ;   

 or something like this, any indicator as filter

Well, I tried to change here..but messing up anything with "Signals" messes up the whole results..

[Maxim Dmitrievsky]

FxTrader562:

Indicator I didn't try yet ..

but I tried " if  (  rand  () /  32767.0  <  0.8  ) "..t hen, it showed random behaviour, because we need to change here also:

if you filter this with MA, it means he will sample buy or sell signals with 0.8 probability when up or dwn trend, so trades will much longer, dont need to change signals filter

I will think how to improve reward func. Tomorrow )

[Maxim Kuznetsov]

Maxim Dmitrievsky:

Better if you change here

if(fast MA>slow MA) if(rand()/32767.0<0.8) return 0; else return 1;

if(fast MA<slow MA) if(rand()/32767.0<0.8) return 1; else return 0;

 or something like this, any indicator as filter

похоже делаем сходные вещи (или пришли к близким выводам)

примерно "если детерминированные алгоритмы проигрывают в условиях недостатка информации (на совершенствующимся рынке), то можно дать волю вероятностям" :-)

[Maxim Dmitrievsky]

Maxim Kuznetsov:

похоже делаем сходные вещи (или пришли к близким выводам)

примерно "если детерминированные алгоритмы проигрывают в условиях недостатка информации (на совершенствующимся рынке), то можно дать волю вероятностям" :-)

тут маленько другое (подготовка размеченных данных для обучения НС), но похоже да

вопрос в том, как псевдослучайным образом лучше размечать данные

[Maxim Dmitrievsky]

FxTrader562:

Ok, let's see...)))

Actually, I am getting pretty impressive results in back-testing so far:))

But the main problem with forward testing:((((

maybe it will never work fine :)

[Maxim Dmitrievsky]

FxTrader562:

No,  Somehow you just the if IMPLEMENT, something to reward  function  related to profits and losses..then closed, the agent of the will of the try to Improve Naturally itself to the make the more profit :)

But it is a big question how to do it :)))

I tried my own way what I showed to you:))

maybe we need samples from another probability distributions,  related to current market situations (distributions), I don't know :)

[Aleksey Vyazmikin]

Куда пропал

Dr. Trader ?