Simbasystem-gbpusd - page 58
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
By the way,some question arises in my mind(occasionaly its happen in my head).
Why we put for TP in Malcik EA value 355? To be close to real trading we have to put 50 or 70 or 90. Could you be so kind to explain briefly,otherwise my head will feel very uncomfortable, because ideas not so often coming there
CheersHi pupsik,
The value 355 is based on the backtests done. How did you figure out the 50 or 70 or 90 setting? What will make you feel comfortable? For me it is serious testing that will find what really is the best setting, not what I feel is one. And testing is exactly how Simba arrived at this value
Obviously, this is just preliminary.
As far as your tests, thanks very much for them. The problem is the quality though. I don't have a charset to display texts in your summary but the percentage number there was 50%-60%. Such quality is basically worthless, we really need no less than 90%. If you have 50%-60%, please stop testing, it's just a waste of your time. Spend your time on getting 90% -- read this thread, this forum, the internet... like I do, hope we will both make it to join the testing team
Malcik
Hi Pilofe,
Thanks for your further replications. There are some minor differences that are small enough to leave us calm and are most likely caused by different data from different brokers. Therefore we can now be glad to be able to start testing full-time.
A couple of days ago, Simba wrote about finding "zones" in Chandelier and TP settings. We need to replicate this too -- we know that the setting found by Simba's Strategy Tester has been replicated BUT the process of finding this setting (i.e. if the setting is really the best one) has not been replicated.
So I wonder if you could please turn on the Optimization feature in Strategy Tester and try to find the best setting for the EA and see if you arrive at the same one as Simba did. Here's his related post:
Unless you have like a NASA supercomputer with a rocket fast CPU, you need to first go in larger steps (e.g. by tens) and once you find such approximate "zone-like" best setting, you can that make it more accurate by going to smaller and smaller steps taken within this smaller and smaller zone you find. For more information on this issue and on optimizing generally, go to Hendrick's thread, first post (https://www.mql5.com/en/forum) where you can download a PDF manual. It is for different EA but the process of optimizing is same in nature.
Please post your conclusions, I'm excited how it turns out. Next step will be the next replication of a Simba's test, and that is W%R setting. Meanwhile, I'll keep trying to get my damn 90% quality (please help, anyone) and I'll also make a 1.2 version of the EA which will be able to test W%R settings. Then we'll wait for comments by Simba who is busy this week.Hi Malcik,
Well,i am not so busy,yesterday I was because I had to go on a business trip,and today,from what I see in the markets I don`t think I will trade until
I get some interesting confirmation in the jpy pairs..;)Chandelier 1h has been broken in usdjpy,eurjpy,gbpjpy..very interesting
Regarding your problem,why you don`t just download a new MT4,login,close it,reopen it -WITHOUT CONNECTION-then do the process again..what I did was (i didn`t delete account,since i am not allowing the testing mt4 to connect) to go to history center double click on m1 ,delete data,import data..close mt4,reopen again-without connecting-file,open offline 1m gbpusd,script,period converter ,5,ok,ok..wait 5 minutes,double click on period converter again..15,ok,ok..wait 5 minutes,etc,etc..complete it until d1 included..when you finish..you close mt4 and reopen again,go to history center and will see that for gbpusd in 1m you have 800k bars,in 5m like 160k,etc,etc..wait before doing this, i will try to email you the 5m files,and,if this works(history center 5m delete,import,etc),i will email all of them sequentially
Hope this helps
Regards
Simba
Pupsik
By the way,some question arises in my mind(occasionaly its happen in my head).
Why we put for TP in Malcik EA value 355? To be close to real trading we have to put 50 or 70 or 90. Could you be so kind to explain briefly,otherwise my head will feel very uncomfortable, because ideas not so often coming there
CheersHi Pupsik,
First ,thanks for your enthusiasm and cooperation..;)
Second,to reply to your question,why don`t you put tp to 50,70,90 ?Test it and compare the results..basically your %profitable trades will increase,but your total profit will decrease..I think that Malcik put those settings,as initial,because from our previous posted tests,these seemed,and still seem to me,the best found settings..
Another thing,try to do the tests on 90% quality data,anything else is misleading and will be a waste of your time,Malcik and myself (and,I presume,hundreds of others)have wasted several hours learning this frustrating lesson,so try to get 90% data quality,and to stick to the testing period(16 june 04 to 23 feb 07)then we can compare results...
Please keep helping us
Regards
Simba
Pilofe
Hi all,
My tests had replicated Simba.
Parameters:
SATL4H,CHAND1H,SATL FILTER:TRUE,ALLOW REENTRIES:TRUE,TP:355PIPS,CHAND DISTANCE 35/100
Thanks...
AngeloHi Angelo,
Thanks..keep posting
Regards
Simba
Hi Malcik,
Well,i am not so busy,yesterday I was because I had to go on a business trip,and today,from what I see in the markets I don`t think I will trade until
I get some interesting confirmation in the jpy pairs..;)Chandelier 1h has been broken in usdjpy,eurjpy,gbpjpy..very interesting
Regarding your problem,why you don`t just download a new MT4,login,close it,reopen it -WITHOUT CONNECTION-then do the process again..what I did was (i didn`t delete account,since i am not allowing the testing mt4 to connect) to go to history center double click on m1 ,delete data,import data..close mt4,reopen again-without connecting-file,open offline 1m gbpusd,script,period converter ,5,ok,ok..wait 5 minutes,double click on period converter again..15,ok,ok..wait 5 minutes,etc,etc..complete it until d1 included..when you finish..you close mt4 and reopen again,go to history center and will see that for gbpusd in 1m you have 800k bars,in 5m like 160k,etc,etc..wait before doing this, i will try to email you the 5m files,and,if this works(history center 5m delete,import,etc),i will email all of them sequentially
Hope this helps
Regards
SimbaHi Simba,
Well that Chandelier on all of these pairs was broken on 2/28 and 3/1 too, it looked like a perfect turning point but it was a loser
But I can see that this could have been very fundamental. Are you going to wait for H4 Chandelier breakout?
What about other pairs (cable, Swiss Frank pairs and aussie)? It looks like all the pairs are now dominated by 205-day cycle...
As far as the data, all that you wrote (incl. the reinstallation) is basically what I did. The most strange thing is that some months ago I had no problem getting the quality, now just can't do it no matter what.
OK, if you could send me the complete M5 file, that would be just great.
Thanks
Optimization
Hi,Simba!
I did 90% testing with latest Malcik EA(he is doing good job!) .
GBPUSD (Great Britan vs US Dollar)
(M5) 2004.06.16 00:00 - 2007.03.02 21:55 (2004.06.16 - 2007.03.04)
MA_Type=1; JMA_HMA_Period=13; JMA_Phase=0; UseSATLFilter=true; AllowReentries=true; TPPips=355; ChandDist_Min=35; ChandDist_Max=100;
The result was 8284.36 $ profit.
I think, because of data.My data from MetaQuotes Software .
Then I played with diffrent setting,to find out the best one.So far I get that one(look to attachment).
Then I decided to optimize Malcik EA.But, computer need 54 hours to optimize it. Now I am thinking - should I do it, do we need it? What do you think, is it worthy?
Pupsik
Hi,Simba!
I did 90% testing with latest Malcik EA(he is doing good job!) .
GBPUSD (Great Britan vs US Dollar)
(M5) 2004.06.16 00:00 - 2007.03.02 21:55 (2004.06.16 - 2007.03.04)
MA_Type=1; JMA_HMA_Period=13; JMA_Phase=0; UseSATLFilter=true; AllowReentries=true; TPPips=355; ChandDist_Min=35; ChandDist_Max=100;
The result was 8284.36 $ profit.
I think, because of data.My data from MetaQuotes Software .
Then I played with diffrent setting,to find out the best one.So far I get that one(look to attachment).
Then I decided to optimize Malcik EA.But, computer need 54 hours to optimize it. Now I am thinking - should I do it, do we need it? What do you think, is it worthy?Hi Pupsik
1-Regarding your tests,it is strange that pilofe could exactly replicate them while you obtained only8k of profit..can you copy and paste(and post
) the settings for your "replication" of my test ?..As I did in post 550 of this thread
2-Optimization:if you can spare 54 hours..it is worth it,if you can`t,don`t worry i will try to do it on the week end.
Regards
Simba
Malcik
Hi Simba,
Well that Chandelier on all of these pairs was broken on 2/28 and 3/1 too, it looked like a perfect turning point but it was a loser
But I can see that this could have been very fundamental. Are you going to wait for H4 Chandelier breakout?
What about other pairs (cable, Swiss Frank pairs and aussie)? It looks like all the pairs are now dominated by 205-day cycle...
As far as the data, all that you wrote (incl. the reinstallation) is basically what I did. The most strange thing is that some months ago I had no problem getting the quality, now just can't do it no matter what.
OK, if you could send me the complete M5 file, that would be just great.
ThanksHi Malcik,
No,I am not going to wait for 4h Chandelier..I already explained the reasoning a few posts ago(was it in this or in the other thread
? )..if there is carry trade resumption ,the move can be fast..so 4h is too slow..if there is carry trade rethinking,there will be a rebound..and then a resumption of uptrend for jpy..so,4h either will not signal anything or will give a false entry at the worst possible moment...I am just waiting for a better cyclic positioning,or,even better a simbasystem 5m long(simultaneous) signal in those 3 pairs that allows for a low risk entry(i don`t count on it happening)..as you said,Chandelier already gave a false 1h signal..but the risk level with chand 1h is acceptable..it doesn`t give many consecutive false signals and the position of conman 4h is much better(for a potential reversal) than a few days ago...
Let`s see what the mirror thinks about my speculation ..;)
other pairs,i am not so focused on them now,I think the money is in the jpy pairs..and I prefer to focus on these 3 "major jpy" pairs..regarding cable ,i think that there is a potential for a rebound today and chf i don`t understand its behaviour,so,i am not following it...
Regards
Simba
Setup for unconventional testing result
Hi Pupsik
1-Regarding your tests,it is strange that pilofe could exactly replicate them while you obtained only8k of profit..can you copy and paste(and post
) the settings for your "replication" of my test ?..As I did in post 550 of this thread
2-Optimization:if you can spare 54 hours..it is worth it,if you can`t,don`t worry i will try to do it on the week end.
Regards
SimbaHello!
This setup give me only 8k per 2,5 years.
Don't be discourage, your's test is valid and for my data I will do optimization and ,I think, will do 44k profit eventually.
//+------------------------------------------------------------------+
//| SimbaSystem.mq4 |
//| Simba, programmed by Malcik |
//| https://www.mql5.com/en/forum |
//+------------------------------------------------------------------+
#property copyright "Simba, programmed by Malcik"
#property link "https://www.forex-tsd.com"
#define MAGICSS 20070129
extern int MA_Type=1; // 1 for SATL, 2 for HULL MA, 3 for JMA
extern int JMA_HMA_Period=13; // When using JMA or HMA, set period here
extern double JMA_Phase=0; // When using JMA, set phase here
extern bool UseSATLFilter=true; // When true, the EA will only go long/short when the price is above/below SATL (or any other MA)
extern bool AllowReentries=false; // When true, more than 1 position can be open (reentries at each 10 pips of better price or after retracement of W%R)
extern int TPPips=355;
extern int ChandDist_Min=35; // Boundaries of distances of price from Chandelier
extern int ChandDist_Max=100;
datetime LastMinute;
int LongTrades=0;
int ShortTrades=0;
double LastBuyPrice=0;
double LastSellPrice=0;
//+------------------------------------------------------------------+
//| expert initialization function |
//+------------------------------------------------------------------+
int init()
{
//----
//----
return(0);
}
//+------------------------------------------------------------------+
//| expert deinitialization function |
//+------------------------------------------------------------------+
int deinit()
{
//----
//----
return(0);
}
//+------------------------------------------------------------------+
//| expert start function |
//+------------------------------------------------------------------+
int start()
{
double SATL; //=iCustom(NULL,PERIOD_H4,"SATL",0,1);
double SATL_Prev; //=iCustom(NULL,PERIOD_H4,"SATL",0,2);
double Chandelier_Up=iCustom(NULL,PERIOD_H1,"ChandelierStops_v1",0,1);
double Chandelier_Dn=iCustom(NULL,PERIOD_H1,"ChandelierStops_v1",1,1);
double WPR=iWPR(NULL,0,90,1);
int res;
//----
if(MA_Type==1)
{
SATL=iCustom(NULL,PERIOD_H4,"SATL",0,1);
SATL_Prev=iCustom(NULL,PERIOD_H4,"SATL",0,2);
}
if(MA_Type==2)
{
SATL=iCustom(NULL,PERIOD_H4,"HMA",JMA_HMA_Period,0,1);
SATL_Prev=iCustom(NULL,PERIOD_H4,"HMA",JMA_HMA_Period,0,2);
}
if(MA_Type==3)
{
SATL=iCustom(NULL,PERIOD_H4,"JMA",JMA_HMA_Period,JMA_Phase,0,1);
SATL_Prev=iCustom(NULL,PERIOD_H4,"JMA",JMA_HMA_Period,JMA_Phase,0,2);
}
LongTrades=0;
ShortTrades=0;
if(OrdersTotal()>0)
{
for(int i=0;i<OrdersTotal();i++)
{
if(OrderSelect(i,SELECT_BY_POS,MODE_TRADES))
if(OrderSymbol()==Symbol() && OrderMagicNumber()==MAGICSS)
{
if(OrderType()==OP_BUY)
{
if(SATL<SATL_Prev || (Chandelier_Up==EMPTY_VALUE && Chandelier_Dn!=EMPTY_VALUE))OrderClose(OrderTicket(),OrderLots(),Bid,3,White);
}
if(OrderType()==OP_SELL)
{
if(SATL>SATL_Prev || (Chandelier_Up!=EMPTY_VALUE && Chandelier_Dn==EMPTY_VALUE))OrderClose(OrderTicket(),OrderLots(),Ask,3,White);
}
}
}
for(i=0;i<OrdersTotal();i++)
{
if(OrderSelect(i,SELECT_BY_POS,MODE_TRADES))
if(OrderSymbol()==Symbol() && OrderMagicNumber()==MAGICSS)
{
if(OrderType()==OP_BUY)
{
if(OrderStopLoss()<Chandelier_Up-Point)OrderModify(OrderTicket(),OrderOpenPrice(),Chandelier_Up-Point,OrderTakeProfit(),0,Green);
}
if(OrderType()==OP_SELL)
{
if(OrderStopLoss()>Chandelier_Dn+Point+Point*4)OrderModify(OrderTicket(),OrderOpenPrice(),Chandelier_Dn+Point+Point*4,OrderTakeProfit(),0,Green);
}
}
}
for(i=0;i<OrdersTotal();i++)
{
if(OrderSelect(i,SELECT_BY_POS,MODE_TRADES))
if(OrderSymbol()==Symbol() && OrderMagicNumber()==MAGICSS)
{
if(OrderType()==OP_BUY)LongTrades++;
if(OrderType()==OP_SELL)ShortTrades++;
}
}
}
if(LastMinute!=Time[0])
{
if(AllowReentries)
{
if(WPR>=-5)LongTrades=0;
if(WPR<=-95)ShortTrades=0;
}
if(LongTrades==0 || (Close[0]<=LastBuyPrice-10*Point && AllowReentries))
{
if(SATL>SATL_Prev && (Close[0]>=SATL || !UseSATLFilter) && Chandelier_Up!=EMPTY_VALUE && Chandelier_Dn==EMPTY_VALUE && Close[0]>=Chandelier_Up+ChandDist_Min*Point && Close[0]<=Chandelier_Up+ChandDist_Max*Point && WPR<=-95)
{
res=OrderSend(Symbol(),OP_BUY,1,Ask,3,Chandelier_Up-Point,Bid+(TPPips*Point),"SimbaSystemM1",MAGICSS,0,Green);
LongTrades++;
LastBuyPrice=Bid;
}
}
if(ShortTrades==0 || (Close[0]>=LastSellPrice+10*Point && AllowReentries))
{
if(SATL<SATL_Prev && (Close[0]<=SATL || !UseSATLFilter) && Chandelier_Up==EMPTY_VALUE && Chandelier_Dn!=EMPTY_VALUE && Close[0]=Chandelier_Dn-ChandDist_Max*Point && WPR>=-5)
{
res=OrderSend(Symbol(),OP_SELL,1,Bid,3,Chandelier_Dn+Point+Point*4,Ask-(TPPips*Point),"SimbaSystemM1",MAGICSS,0,Red);
ShortTrades++;
LastSellPrice=Bid;
}
}
LastMinute=Time[0];
}
//----
return(0);
}
//+------------------------------------------------------------------+
Respect.
Getting 90% quality.
Hi Simba,
Well that Chandelier on all of these pairs was broken on 2/28 and 3/1 too, it looked like a perfect turning point but it was a loser
But I can see that this could have been very fundamental. Are you going to wait for H4 Chandelier breakout?
What about other pairs (cable, Swiss Frank pairs and aussie)? It looks like all the pairs are now dominated by 205-day cycle...
As far as the data, all that you wrote (incl. the reinstallation) is basically what I did. The most strange thing is that some months ago I had no problem getting the quality, now just can't do it no matter what.
OK, if you could send me the complete M5 file, that would be just great.
ThanksHi,Malcik!
Thank you for your hard work! I am happy to use your EA.
I can suggest to check at Metatrader Service - Setting, or just do Ctrl+O,and in Setting click on Graphik( I have russian Metatrader, mabe translation wrong). In Graphik chose value of Max. bars in window 1000000. It will slow your comp, but you can get 90% quality.
If it gonna help - I am happy, if not - I give a try.
With respect, Pupsik.