I can't believe it! - page 7

 
sllawa3 >> :

I have e.g. entry above the bar opening level as I understand the author's too.

Above (below) by a certain delta in pips? Which pair are you testing?

 
sllawa3 >> :

tried it says it's too big...

By the way about the modifications I want to say that not so much... (>> not on every tick and only when the price level exceeds the previous one by a pip.)

try archiving (in RAR or Zip)

But what about modifications: On real account it will not work like that (like with xnko), you will get a ban from your broker. But there is a way out - to "modify" virtually...

 
Vinsent_Vega >> :

try archiving (in RAR or Zip)

My only problem is not solving one question, it's the closing condition when the bar changes ( why I trade semi-automatic). But there is a way out - to "modify" virtually...

i have no problems with the modification i have only one unresolved issue, the closing condition when the bar changes ( that's why i trade in semi-automatic mode )

I do not know how to write a condition that if the order at bar change in loss - was closed

 
sllawa3 >> :

I have no problem with the modification I have only one question to solve - the closing condition when the bar changes (that's why I trade in semiautomatic mode)

What's the problem? If the bar is not the first to open the order, let's close it.

if (iBarShift( symbol, timeFrame, OrderOpenTime()) > 0)
{
   OrderClose(...);
}
This is in the order loop at the start of each bar.
 

Well looks like it's time to stop drooling once again)

Test EUR/JPY from 01.01.08 to 08.02.08

1. with a modelling quality of 44%, profit 40 000

2. 90% modeling quality loss 10 000 (equal loss)

 
Dezil >> :

Well looks like it's time to stop drooling once again)

Test EUR/JPY from 01.01.08 to 08.02.08

1. with a modelling quality of 44%, profit 40 000

2 at 90% modeling quality 10,000 loss (equal to loss)

Yeah... I can't do it on automatic... although the system works fine on semi-automatic...

 
sllawa3 >> :

I do not know how to write the condition that if the order at the change of the bar in the loss - to close

Check the formation of a new bar using a static variable and close - if OrderProfit () < 0, then - OrderClose (). Approximately like this:

static datetime prevtime;
...
if(prevtime != Time[0])

{
prevtime = Time[0];
if (OrderProfit () < 0)
OrderClose();
}


 

What is the deeper meaning of closing a position at the end of a bar? If the position is profitable, then it makes sense to close it to the end.

My Expert Advisor does not close positions at the end of a bar, and it does not matter (it is even worse when the bar is closed). My point is that the difference in the testing quality is crucial.

I advise the author to achieve the modeling quality of 90%

Although, judging by his disappearance, he has achieved it)

 
Dezil >> :

Well looks like it's time to stop drooling once again)

Test EUR/JPY from 01.01.08 to 08.02.08

1. with a modelling quality of 44%, profit 40 000

2 with 90% modeling quality loss 10 000 (equal loss)

it may be time to call it a day, but it's an interesting algorithm to work with...

 
Vinsent_Vega >> :

>> it might be time to stop with the EA, but it's an interesting algorithm to work with...

I tried it some time ago, but nothing good came out of it. If you have fresh ideas, welcome, as they say)