Development plans for the MetaTrader 5 Strategy Tester - page 20
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
Tried to do Optimisation by real ticks without filtering. For this I had to disable RAM-Drive and work with the tester via SSD.
SSD blinks all the time during optimization. Some wild activity from the Tester's side. This despite the fact that it takes 30 seconds for each pass.
These Agent\temp\bar*.tmp files are several gigabytes in size for what? Why read them all the time during Optimization?
Did you get an answer about all these tmp files?
Did you get an answer about all these tmp files?
No.
For those testing, especially on broker history, the "exclude repetitive ticks" feature would be very useful (e.g. make it next to "profit in pips to speed up calculations")
On one popular broker I found that 8mln ticks out of 13mln per month are repetitive! Thus, we can significantly increase the speed of testing for purchased EAs or those that do not have such a program filter.
I also ask to make it possible to select more column parameters on the optimization results page. For example, I want to see the drawdown in the deposit currency during optimization with a fixed lot value, but it is impossible to select it - onTester is occupied by another parameter.
Dear developers, the following indicator would be very useful in the optimization results:
Equity Drawdown Relative- the largest percentage drop in equity between the local maximum and the next local minimum.**The optimizer now givesthe Equity Drawdown Maximalas a percentage-the largest drawdown of funds between the local maximum and the next local minimum in the deposit currency.
Example from the reference:
The largest drawdown percentage was 33.3%, but the optimizer will show a drawdown percentage of 31.25%(Maximal Drawdown). Consequently, if the balance of the optimizer is growing, the last drawdown in % (if it is falling down, the first drawdown in %) will be shown rather than the maximal drawdown in % for the whole testing period.
The largest drawdown percentage was 33.3%, but the optimizer will show a drawdown percentage of 31.25% (Maximal Drawdown). Consequently, with a growing balance we will most likely see the last drawdown in % (with a declining balance we will see the first drawdown in %) and not the maximal drawdown in % for the whole testing period.
I wrote about it recently, there was no reply:
Forum on trading, automated trading systems and testing trading strategies
Maximum and Relative Drawdown. Mt5 Tester
Andrey Khatimlianskii, 2021.03.10 20:24
I was surprised to find out that the "DrawDown %" column in the optimization results shows the percentage value corresponding to the maximal drawdown in money. Which is far from always the maximum drawdown percentage.
Was it intended that way? It would be much more useful to see the maximal drawdown percentage value (STAT_EQUITY_DDREL_PERCENT).
Here is an example of results where the maximum drawdown in money (4077.65) as a percentage (26.72%) is much less than the maximum relative drawdown (3795.43 = 35.61%):
Here is how it looks on the chart: approximately the same drawdowns (in money terms) occurred at different balances:
The optimisation table shows a figure of 26.72:
With a fixed lot these numbers would be the same, but if dynamic money management is used, relative drawdown should take precedence, in my opinion.
I have, of course, added the castum criterion (already calculated and displayed in the screenshot above), but that doesn't solve the problem of using the wrong drawdown when calculating other criteria.
Consider replacing or adding a new column with Relative DD?
I wrote about this recently, there was no response:
In Excel, you can dump a table from all the parameters in an opt file.
A table of all parameters in the opt-file can be dropped into Excel.
It is possible to calculate your criterion as well.
This is a boxed solution, it is not clear what the reason is for this choice.
Why do you think there are so many profits at 6pm? The answer lies in the fact that any scalper can be modified so that such a strong surge of profit will occur at a predetermined hour.
I.e. there is rather little use in the current algorithm for building this chart. I think many people know the answer to the question.
I noticed that when I work with the MT5-Tester, I do 95% of my interactions with it only through the mouse.
That is, a little less than completely, it is geared to work through the mouse.
ZZY Switching between Tester tabs should be done using hotkeys.
I've been trying to make a custom test of my strategy. The other day I tried to do a custom testing of my strategy and faced the same problems, about which I wrote earlier and which solutions were suggested by fxsaber(here) and Francuz(here).
If to summarize what has been said, from an application point of view the required improvements are quite simple:
1. In the functionality of Services and Expert Advisors, add a new function OnTesterInit(), which is called before OnInit() in case a Service/Advisor is launched in the Strategy Tester.
2. Within the OnTesterInit() function, make it possible to set up testing using 3 key functions:
- TesterSetInfo() - for automatically setting the start/end dates, character set and other basic testing variables,
- TesterSetCharts() - for automatic setting of necessary charts and application of saved templates to them,
- TesterSetExpert() - for automatically setting a set of variables of the tested Expert Advisor and setting the Expert Advisor to be tested when called from the Service.
This will completely cover the tasks of test automation, both in the form of a sequence of "tasks" and in the form of numerous runs on the custom logic.
3. we also need to provide the operation of the custom EventChartCustom events in the Tester.