Multicurrency expert test results - page 3

New comment
 
stringo:

What's wrong with opening the next bar then?

By the way, I wrote a trading system in MQL4 so that no matter what testing mode is used (by opening prices, reference points, all ticks), the results were identical. Therefore, after making sure of that, I tested and optimized by opening prices. And that mode is many times faster there than in MT5. Very significant. Can you make another mode by opening prices like in MT4? ))) When visually testing, you can see that in the open price mode in MT5 the bar is formed by ticks (or checkpoints?). It's so long...
 
tol64:

No. The phrase "Test on EURUSD instrument from GBPUSD chart" means that you are trading onEURUSD, but the Expert Advisor is onGBPUSD chart. All results are onEURUSD, I just switched from one symbol to another.

So, eitherInteresting's assumption is not confirmed, or you have something wrong.

Pay attention to the part of the code:

int OnInit()
{
if(iCustom("EURUSD",PERIOD_D1,"Spy Control panel MCM",ChartID(),0,CHARTEVENT_TICK) == INVALID_HANDLE)
   { Print("Ошибка установки шпиона на EURUSD"); return(true);}
   
if(iCustom("GBPUSD",PERIOD_D1,"Spy Control panel MCM",ChartID(),1,CHARTEVENT_TICK) == INVALID_HANDLE)
   { Print("Ошибка установки шпиона на GBPUSD"); return(true);}
}
Here you can see that you are "hinging" some "Spy Control panel MCM" indicator on two different characters. That is, you have different symbols as signal sources. But you claim that "we trade onEURUSD", i.e. the source of the signal is one and the same symbol. Let's make it clear.
 
tol64:

Of course, this also has to be taken into account. In a programme designed for automatic trading, everything has to be taken into account. At least as much as possible. At the moment there are two modes: Normal and Arbitrary delay. The developers have already announced that they will gradually bring the testing process closer to reality. That's encouraging. Illusions need to be cut at the root.

No, my point is slightly different. Imagine that the Expert Advisor receives a series of certain signals from "spies". The Expert Advisor starts to process the first signal, but what happens to the others?
Yedelkin:

So eitherInteresting's assumption is not confirmed, or you have something wrong.

Pay attention to the part of the code:

Here you can see that you "hinge" some "Spy Control panel MCM" indicator on two different symbols. That is, you have different symbols as signal sources. But you claim that "we trade onEURUSD", i.e. the signal source is one and the same symbol. Let's make it clear.

Judging by this code, the signals do come from two characters, except that the Expert Advisor may process one of these signals with a delay.

 
stringo:

How then is the opening of the next bar inappropriate?

Because the opening of bars on different instruments is at different times. And the closing is exactly on the clock, the same for all.

In multicurrency testing, there is a significant difference. There are even examples at the beginning of this thread.

 
tol64:

All in all. Interested in everyone's opinion and even more interesting in the solution methods.))

I'm using OnTick and making a check for a per-bar synchronisation across all the working tools.
 
MetaDriver:

With multi-currency testing there is a significant difference. There are even examples at the beginning of the thread, very illustrative.

This is not serious. A single tick cannot make a significant difference. There's just something wrong with it.
 
Yedelkin:

So eitherInteresting's assumption doesn't hold true, or you have something wrong.

Pay attention to the part of the code:

Here you can see that you "put" some "Spy Control panel MCM" indicator on two different symbols. That is, you have different symbols as signal sources. But you claim that "we trade onEURUSD", i.e. the signal source is one and the same symbol. Let's define it.

Oh, that's getting closer. There seems to be one option on which I'm wrong.)) I'll think about it for a moment and write more about it straight away...

Interesting

No, I'm talking a little bit about something else. Imagine that the Expert Advisor has received a series of certain signals from "spies". The Expert Advisor starts to process the first signal, but what happens to the rest?

Likewise.) The right questions are sometimes the answers. I have a long test going on right now. A little later, when the test is over I will try one guess. I'll let you know the result. Thank you.
 
TheXpert:
That's not serious. One tick can't make a significant difference. There's just something wrong with it.

"- Defendant, why did you kill the old lady? She only had twenty kopecks!

- Well... One old lady is twenty kopecks, the second old lady is twenty kopecks... five old ladies is a ruble..."

 
MetaDriver:

"- Defendant, why did you kill the old lady? She only had twenty kopecks!

- Well... One old lady is twenty kopecks, another old lady twenty kopecks ... five old ladies is already a ruble..."

I would like to present an instructive and illustrative example. I'm not going to write the name of the program, I think the rules prohibit it, but I can think of pictures)).

One and the same simple trading strategy. But different spread and slippage:

Blue shows perfect result without spread and slippage. Yellow shows the result with a spread of 2 pips and slippage of 10 pips on each trade. And the third result is lilac Equity, it shows a result with a 5-point spread and slippage of 15 points on each trade.

By the way, I would like the developers to pay attention to the display of Equity in the forward testing mode. It is very clear! Please take note of it. Please support forum participants, if interested, of course)).

 
marketeer:
I'm using OnTick and I'm doing a check for a per-tick synchronisation across all the working tools.
Could you show an example similar to what I did at the beginning of the thread. Enable trading on only one symbol, but test the EA on the other. Take a screenshot and test it on the symbol on which the test is running. Will there be divergence like that shown at the beginning of the branch? Although if the bar formation is tracked on all symbols, it should be identical. But still you need to check everything...
1234567
New comment