Do you have any tactics for dealing with the loca? - page 62

New comment
 
Stells:

Good afternoon.

Igor, it seems to me that your so called MM is the same EA, and it should be more profitable than the EA where your function is added to win back losing positions.

The question is why don't you use your MM as an independent EA?


finally a bright idea.... - Yes you are right it is too modified avalanche, from avalanche remained only the idea of lock on lock ..., even the concept of "corridor / price channel" has changed, based on the current market volatility, rather than the distance for the next lock

I would never launch an EA as an independent one, it is based on constant increase of lots to achieve profits, but at the same time, if we keep control of available funds before setting another lock, the results are quite "fantastic".i.e. it is better to make a profit within the spread if the price starts to go back without reaching ТР than to take "losses" holding my avalanche high, i wrote about it - if i make a loss (i set 300$ in fixed numbers) then let's start an avalanche or something like that

 
Igor, could you please elaborate a bit more on this "compulsory equity trawl", I don't quite understand how such a trawl should work...
 
renoshnik:
Igor, could you please elaborate a bit more on this "compulsory equity trawl", I don't quite understand how such a trawl should work...
/жжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжж
//+------------------------------------------------------------------+
//|Трал по эквити                                                    |
//+------------------------------------------------------------------+
/*
Функция вызывается в самом начале эксперта сразу за int Start(), вызов происходит таким образом :

//удаляем запрет на торговлю после удаления всех ордеров 
  if(OrdersTotal()<1){
    if(GlobalVariableCheck("stop")){GlobalVariableDel("stop");}  
    gEqviti=AccountBalance(); 
  } 
//тралим эквити
  if(EcvitiTral3(EqwTralStep)){return(0);}
  if(GlobalVariableCheck("stop")){return(0);}

если еквити поднялась над балансом, то она блокирует все стальные функции эксперта и тралит.
теперь по переменным : в глобальных переменных static double gEqviti;
во внешних переменных extern double EqwTralStep=0.03; 
шаг трейлинга в процентах от эквити extern bool WithoutLoss=false; разрешение на применение метода безубытка
работа фунции : при превышении эквити над балансом она запоминает уровень баланса как нулевой, 
при котором надо закрывать- это работа с безубытком, при дальнейшем превышении эквити 
на размер EqwTralStep в прцентах она передвигает уровень закрытия выше, 
если скорость превышения большая то функция увеличивает шаг квадратично .

*/
//жжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжжж
bool EcvitiTral3(double EqvTralStep){
//if(GlobalVariableCheck("stop")){GlobalVariableDel("stop");}  
if(OrdersTotal()<1){gEqviti=AccountBalance();return(false);}

   if(WithoutLoss){
    if(AccountEquity()>=AccountBalance()){mlots = 0.01;CloseAllFirstProfit();
       WithoutLoss=false;Print("CloseAll");GlobalVariableDel("stop");return(false);}
    return(true);
   }
   
   if(AccountEquity()>AccountBalance()){
      if(AccountEquity()-AccountBalance()>((AccountBalance()/100)*(EqvTralStep*2))){EqvTralStep=EqvTralStep+EqwTralStep;}      
      if(AccountEquity()>(gEqviti+(gEqviti/100*EqvTralStep))){
        gEqviti=gEqviti+(gEqviti/200*EqvTralStep);
        if(!GlobalVariableCheck("stop")){GlobalVariableSet("stop",0);}
      }
      if(AccountEquity()<=gEqviti){mlots = 0.01;CloseAllFirstProfit();
         Print("CloseAll");GlobalVariableDel("stop");return(false);}
      else{Comment(AccountFreeMargin()+AccountMargin(),"\n",
                 "EQUITY TRALING MODE\n",
                 "EQUITY TRALING STEP   =",EqvTralStep," %",
                 "\nACCOUNT BALANS         = ",AccountBalance(),
                 "\nCLOSE LEVEL                  = ",gEqviti,
                 "\nACCOUNT EQUITY        = ",AccountEquity(),
                 "\nNEXT STEP                     = ",gEqviti+(gEqviti/100*EqvTralStep));
               //  Print("AccountEquity() = ",(AccountBalance()/100)*(EqvTralStep));
                 return(true);
      }
    }else{GlobalVariableDel("stop");}  

return(false);
}

The code is not mine, I haven't even understood it yet, but it works 100%.
 
renoshnik:
Igor, could you please elaborate a bit more on this "compulsory equity trawl", I don't quite understand how such a trawl should work...
The equity trawl is not obligatory - after entering the profit zone, you can close the overlapped ones with the same effect and trawl the remaining order with an ordinary trawl.
 
khorosh:
Trawling over equity is not necessary. After entering the profit zone, you can close overlapped orders and trawl the remaining order with the same effect.


Perhaps, but so far I had the idea to make a universal function for "pulling" in positive results of losing EAs, while tral on equity is just less of a hassle - no need to take into account the 4 (5) sign, minimal stops, etc. - Just plug in the function and that's it

ZS: Well, what's the idea? renoshnik you have another EA in kodobase which doesn't want to work in your tester, i don't have time today, but you can redo sllep() in the EA like this

      ..................
      datetime opentim=OrderOpenTime();
      ........
// в start()
      if ((TimeLocal()-opentim)<1000) return(0);

about this way you can make sleep() for one second

I think this EA will show better results :)

 
IgorM:


SZS: Well, how's the idea? renoshnik you have another EA in kodobase which does not want to work in your tester, I don't have time today, but you can re-do sllep() in the EA like this:

you can do sleep() for one second or so

I think this EA will show great results :)

I solved this problem, http://voloshin-fxcci.blogspot.com/ my Expert Advisor has been working on real account since January. So far (don't jinx it) every month is in profit... By the way, here is a snapshot from the beginning of the month, even "lots" (an averaging order) managed to work...

I tried to do something "avalanche-like" in this EA, but I don't know yet how to do it.

I think that equity trawl will not work for this EA in particular. As far as I understood, equity trailing is applied when only ONE position is opened and I have many positions with some drawdown due to "noise" which I am not going to deal with waiting for the price to reverse. But among all these positions there is ONE or TWO that have sagged to a "critical" level and there is no expectation that price will be able to roll back that far. It is only these few positions that need to be "locked" - then an equity trawl will not work....

 

Found this in the codebase archives == Introducing the Anti-Moose programme

 
renoshnik:

I found this in the archives of codebase == Introducing the Anti-Moose program


I just started building my "Anti-Loss" today. If the avalanche aims to earn on lots, then it makes more sense for Anti-Loss to return to zero and the total volume of all open positions in each direction has to be recalculated, while the tester shows it that way

there is no trawl on equity at all, we simply overlap at a drawdown of x points. but this is a problem when there are many open orders, as the increase in the overlap is minimal

 
IgorM:


I just started building my "antilossoy" today if the avalanche aims to breakeven, then it makes more sense for the antilossoy to return to zero and the total volume of all open positions in each direction should be recalculated, so far it is like this in the tester

I have just started making my "antilosse" today. if the avalanche aims to earn from lots, then it is more logical to withdraw them to zero and it is imperative to recalculate the total amount of lots.


May I make a small suggestion... If we are going to do "antilope", we think it should be a script and it would be good if when we run this script we could set a "ticket" for the order which should be taken to Breakeven without affecting other orders ....
 
renoshnik:

Can I make a small suggestion... If you want to make an "antilope", you would have to be a script and it would be nice if when you start the script you could set a "ticket" for an order to be taken to Breakeven without affecting other orders .....


half a day "twisting the tester" - the conclusion is one: lock is the application of another strategy to a strategy that gave losses at a certain point in time, in combination with an effective strategy lock increases the efficiency of the EA several times, but.....

- lock should be placed on a different strategy/indicator

- learning locking "out of the blue" (without a strategy) is inefficient

- Placing stop pending orders is not a solution, because the level of a stop order must be calculated depending on the instrument volatility.

But as a result - using my strategy of locking of random losses I observe 5 "reversals" on the history so far, while I try to increase the volume of the next lock, but the connection between time and money is clearly seen, the higher the increase in the lock-up ratio, the faster it will unravel, but the higher the drawdown

Drawdown is less in this case, since I've changed to arithmetic progression of lock volume increase depending on the volumes already exhibited on the market

1...55565758596061626364
New comment
Reason: