Experts: e-PSI@MultiSET

New comment
 

e-PSI@MultiSET:

On a way to Grail!...

Author: TarasBY

 
  1. Average = sum / count
    fda_AvarageCandle[0] /= (li_cnt_Bar - 1);
fda_AvarageCandle[1] /= (li_cnt_Bar - 1);
  2. int li_pips = li_cmd * (OrderClosePrice() - OrderOpenPrice()) / gd_Point;
    On a 5 digit broker a pip is not a point
    //++++ These are adjusted for 5 digit brokers.
int     pips2points;    // slippage  3 pips    3=points    30=points
double  pips2dbl;       // Stoploss 15 pips    0.015      0.0150
int     Digits.pips;    // DoubleToStr(dbl/pips2dbl, Digits.pips)
int     init(){
     if (Digits % 2 == 1){      // DE30=1/JPY=3/EURUSD=5 forum.mql4.com/43064#515262
                pips2dbl    = Point*10; pips2points = 10;   Digits.pips = 1;
    } else {    pips2dbl    = Point;    pips2points =  1;   Digits.pips = 0; }
    // OrderSend(... Slippage.Pips * pips2points, Bid - StopLossPips * pips2dbl

  3. if (DayOfWeek() == ar_BadDay[li_DAY])
    forum.mql4.com/33851 reports DayOfWeek() always returns 5 in the tester. I only use the Time* versions. TimeDayOfWeek(Time[0])
  4. extern string BadDay  = "0"; // Плохие дни недели для работы (через ",")
extern string BadTime = "0"; // Плохое время для работы (через ",")
    While your version allows multiple times and days, they can't be optimized in the tester. My version allows start/stop time and multiple days.
 
WHRoeder:
  1. Average = sum / count
    fda_AvarageCandle[0] /= (li_cnt_Bar - 1);
fda_AvarageCandle[1] /= (li_cnt_Bar - 1);
  2. int li_pips = li_cmd * (OrderClosePrice() - OrderOpenPrice()) / gd_Point;
    On a 5 digit broker a pip is not a point
    //++++ These are adjusted for 5 digit brokers.
int     pips2points;    // slippage  3 pips    3=points    30=points
double  pips2dbl;       // Stoploss 15 pips    0.015      0.0150
int     Digits.pips;    // DoubleToStr(dbl/pips2dbl, Digits.pips)
int     init(){
     if (Digits % 2 == 1){      // DE30=1/JPY=3/EURUSD=5 forum.mql4.com/43064#515262
                pips2dbl    = Point*10; pips2points = 10;   Digits.pips = 1;
    } else {    pips2dbl    = Point;    pips2points =  1;   Digits.pips = 0; }
    // OrderSend(... Slippage.Pips * pips2points, Bid - StopLossPips * pips2dbl

  3. if (DayOfWeek() == ar_BadDay[li_DAY])
    forum.mql4.com/33851 reports DayOfWeek() always returns 5 in the tester. I only use the Time* versions. TimeDayOfWeek(Time[0])
  4. extern string BadDay  = "0"; // Плохие дни недели для работы (через ",")
extern string BadTime = "0"; // Плохое время для работы (через ",")
    While your version allows multiple times and days, they can't be optimized in the tester. My version allows start/stop time and multiple days.
Everything is working as intended. And if you do something do not understand - ask!

And a solution can always be a few - and all at once right!

Exemple:

    fda_AvarageCandle[0] /= (li_cnt_Bar - 1);
    fda_AvarageCandle[1] /= (li_cnt_Bar - 1);

because for

    for (int li_BAR = 1; li_BAR < li_cnt_Bar; li_BAR++)

How many elements between 1 and (li_cnt_Bar - 1)???

Digits 4 or 5 is calculated here:

    if (gi_Digits == 3 || gi_Digits == 5)
    {gi_Decimal = 10;}
    TakeProfit *= gi_Decimal;
    gd_TP = TakeProfit * gd_Point;
    StopLoss *= gi_Decimal;
    gd_SL = StopLoss * gd_Point;
    ProfitMIN *= gi_Decimal;
    Dopusk *= gi_Decimal;
    gd_dopusk = Dopusk * gd_Point;
    Slippage *= gi_Decimal;
    gd_step = 1 * gi_Decimal * gd_Point;

if (DayOfWeek() == ar_BadDay[li_DAY])
Not tested, but to correct.

        if (TimeDayOfWeek (gdt_curTime) == ar_BadDay[li_DAY])
 

Hi, Thanks for sharing. Always worth the effort to get as good a modelling as possible. 99% tells a different story on EURUSD H1.

There is a good article at http://eareview.net/tick-data and several others if you google for 99% MT4 backtesting.

Cheers

Brooky

 
brooky29:

Hi, Thanks for sharing. Always worth the effort to get as good a modelling as possible. 99% tells a different story on EURUSD H1.

There is a good article at http://eareview.net/tick-data and several others if you google for 99% MT4 backtesting.

Cheers

Brooky

Settings (default) are optimized for a specific TF (M1: NewBarInPeriod = 1; ControlPeriod = 60), a specific Symbol (EURUSD), for a specific broker. Illiterate initiative is doomed! Read carefully the annotation!
This version of the advisor in the tester NOT OPTIMIZE!

This is (currently Advisor) is only one part of the whole system.

P.S. NEVER work "at every tick" - my personal Opinion!

 
TarasBY:
brooky29:

Hi, Thanks for sharing. Always worth the effort to get as good a modelling as possible. 99% tells a different story on EURUSD H1.

There is a good article at http://eareview.net/tick-data and several others if you google for 99% MT4 backtesting.

Cheers

Brooky

Settings (default) are optimized for a specific TF (M1: NewBarInPeriod = 1; ControlPeriod = 60), a specific Symbol (EURUSD), for a specific broker. Illiterate initiative is doomed! Read carefully the annotation!
This version of the advisor in the tester NOT OPTIMIZE!

This is (currently Advisor) is only one part of the whole system.

P.S. NEVER work "at every tick" - my personal Opinion!

Well I must say I have never been called Illiterate before.

I obviously have no idea what I am doing after all these years http://www.Brooky-Indicators.com

As usual I offered the 99& modelling link as a possible assistance in you achieving your goals as I have offered many of my indicators for free which have been downloaded over 30,000 times.

Good luck with your endevours.

Cheers

Brooky

 

brooky29:

Well I must say I have never been called Illiterate before.

I obviously have no idea what I am doing after all these years http://www.Brooky-Indicators.com

As usual I offered the 99& modelling link as a possible assistance in you achieving your goals as I have offered many of my indicators for free which have been downloaded over 30,000 times.

Good luck with your endevours.

Cheers

Brooky

So we have different opinions. It happens ... ;)
Good luck.
 
Ezis:

hi, there are modifed Ea And profitable codebase.mql4.com/5762
Administrator! What kind of hidden advertising is distributed on the site!?
 

For enthusiasts of extreme!

I do not meant to put this version of the advisor on several instruments simultaneously on a single account and, accordingly, global variables, this Advisor does not have a specific prefix that with the joint work of several advisers made ​​not predictable consequences in the final result...
 
TarasBY:

For enthusiasts of extreme!

I do not meant to put this version of the advisor on several instruments simultaneously on a single account and, accordingly, global variables, this Advisor does not have a specific prefix that with the joint work of several advisers made ​​not predictable consequences in the final result...


Can you make a reverse version or a reverse option of your e-PSI@MultiSET.mq4, now when the market breaks out up, it places sell orders and you have to wait until

the market comes back and if not you will lose, why don´t go with the market and run two version of your e-PSI the original (a taste of martingale) and parallel

the reverse version. I am not a coder, if you can change your code, it yould be great.

 
fixfoxi:
TarasBY:

For enthusiasts of extreme!

I do not meant to put this version of the advisor on several instruments simultaneously on a single account and, accordingly, global variables, this Advisor does not have a specific prefix that with the joint work of several advisers made ​​not predictable consequences in the final result...

Can you make a reverse version or a reverse option of your e-PSI@MultiSET.mq4, now when the market breaks out up, it places sell orders and you have to wait until

the market comes back and if not you will lose, why don´t go with the market and run two version of your e-PSI the original (a taste of martingale) and parallel

the reverse version. I am not a coder, if you can change your code, it yould be great.

I did't understand what "feature" of your proposal. WHAT reverse??? DO NOT useless work, sorry!
12
New comment
Reason: