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
Dear developers, please make the optimiser and tester work until the current time, not until the end of the previous trading day.
have you tried setting the end date a day ahead?
I am...
is it working?
No, of course not! I would be here. in fact, this problem has already been discussed by me with the developers in this thread...
it was explained that most people do not understand why test results differ when tested on the same day period.
and following the principle "but don't let anyone get you..." they just limited the testing period on the right to the end of the previous trading day, which is wrong.
Dear developers, please make the optimizer and tester work until the current time, not until the end of the previous trading day.
Now the situation is such that the current trading day is completely out of the optimization period (((.
Many systems are running on timeframes much less than D1 and this is critical for them!
We will not do it for two reasons:
1. unnecessary network activity every time we run a test, as the data for the current day is constantly changing. Could be critical for cloud agents.
2. Potential non-repeatability of results from pass to pass. That is, on optimization you will get one result, and on a single run, oppa, another. As the data for the current day is constantly changing.
We will not do this for two reasons:
1. unnecessary network activity every time we run a test, as the data for the current day is constantly changing. For cloud agents, this could be critical.
2. Potential non-repeatability of results from pass to pass. That is, on optimization you will get one result, and on a single run, oppa, another. Since data for the current day is constantly changing.
1. limit such texturing to local agents only.
2. Limit optimization to exclude the current day.
So, current day test will be available only on local agents and not in optimization.
The idea is that you can compare real and tester inputs right now, and not wait for the next day.
We will not do this for two reasons:
1. unnecessary network activity every time we run a test, as the data for the current day is constantly changing. For cloud agents, this could be critical.
2. Potential non-repeatability of results from pass to pass. That is, on optimization you will get one result, and on a single run, oppa, another. As the data for the current day is constantly changing.
no one is asking for relevance to the second... but one trading day is, sorry, out of the question.
points 1 and 2 go away if you fix the right time at the start of the test/optimisation.
no one is asking for relevance to the second... but one trading day is, sorry, out of the question.
Then you will have to enter not only the year month date, but also the hour:minutes end.
Tester has possibilities to do optimisation on 10 year history, think 1 trading day is a grain of sand.
If you don't want to trade on one tick, just create a "tester" and use it to finish the last trading day (thanks, MQL5 has tools for that).
Is this a bug or a fix?)
Then you will have to enter not only the year month date, but also the hour:minutes end.
The tester has the ability to do optimisation on a 10 year history, I think 1 trading day is a grain of sand.
I don't see any problem with it, if you don't have a real trading history and you don't want to waste it, just use "tester" program and run it on last trading day (thanks MQL5 has tools for that).
You see, what's important in trading on financial markets is what's happening now...
and it almost doesn't matter what happened 10 years ago.
you see, in trading the financial markets, what's important is what's happening now... it almost doesn't matter what happened 10 years ago... one last day of trading is very important.)
i agree that there are systems where this is not very critical, but also desirable.