Multicurrency expert test results - page 7
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
So 15 minutes to pass the test is over the top, even on 12 currencies.
You read my post again. During 14.65 minut 3489 trades the Expert Advisor places market orders, modifies and closes them, places pending orders, modifies and closes them. I have 6 functions doing this. I have not added that all these trading functions have auxiliary functions which also contribute to their performance. Do you thinkthis functionality is not enough for most strategies?
If you don't take history into account, don't optimise and sacrifice neural network functionality.
In other cases, you have to keep in mind the heavy indices and mathematics.
Alexander, you wrote here once:
papaklass 2010.09.09 16:24 #
I've reworked the algorithm a bit and the test runs for 727 seconds on 12 currencies. Optimise algorithm.
By "Optimize the algorithm" did you mean to optimize the algorithm that Nikolay Kositsin suggested?
True, before your post Nikolay wrote:
GODZILLA 2010.09.09 14:18 #
You can also simply use the OnTimer() function. Well, in terms of code optimization there are still inexhaustible possibilities.
Alexander, thank you for such a thorough explanation. I really like the idea of tools being processed on more than one tick. I will definitely try this variant. The second and third items are implemented in the same way, but more elegantly than mine.
Of course, many thanks to the authors of articles and company MQ for their efforts!
papaklass:
I hope this conclusion is drawn in relation to an individual strategy yes on a specific period, i.e. without generalising to the results of all other experts.
Please read my post again. For 14.65 minut 3489 trades and at the same time the Expert Advisor places market orders, modifies and closes them, places pending orders, modifies and closes them. I have 6 functions doing this. I have not added that all these trading functions have auxiliary functions which also contribute to their performance. Do you think this functionality is not enough for most strategies?
I really like the idea of having the tools processed on more than one tick. I will definitely try this option.
I do not see any convincing arguments for using such a scheme in a multicurrency algorithm, except for artificial shortening of testing time for passing championship tests.
In the real world, one tick is enough for all instruments, considering that there will not be any operations with many symbols. But testing results may significantly differ in comparison with 'one tick - all instruments'. Check it in your tests (it depends on the Expert Advisor).
Valmars:
Не вижу никаких убедительных доводов для использования такой схемы в мультивалютнике, кроме искусственного сокращения времени тестирования для прохождения тестов чемпионата.
В реале и одного тика на все инструменты хватит за глаза, учитывая, что по многим инструментам не будет операций. А вот результат тестирования может существенно различаться по сравнению с 'один тик - все инструменты'. Проверьте в своих тестах (зависит от эксперта).
Tried it. The duration of the tests did not improve significantly. It was ~27 minutes, now it is 25-26 minutes on 12 instruments. There are no questions about the Expert Advisor's work in real time. Everything works well with all methods listed above.
Generally, the matter of testing Expert Advisors is very delicate. I have not made up my mind yet. As for the championship, I've already noticed some participants wrote that the speed of tests is 3-4 times faster on the Organizer's equipment. I haven't sent my EA for automatic testing yet, there are still some tests to be added. Hopefully 27 minutes will turn into 10. ))