Expert profiling conundrum - how not to exceed the 15 minute limit of the Championship
Any ideas?
This is an interesting situation. My multicurrency Expert Advisor, unencumbered by any online retraining, does not fit into the 15 minute limit.
I have inserted a primitive profiler, overlaying my code calls with it. It turns out that my code works for only 1.5 minutes during the test since the beginning of the year which takes 16 minutes. Where does the rest of the time go? And how can I optimise something which is not in EA?
Do you have any ideas?
Indicators are likely to be a bottleneck (if there are any). Further you need to look at more details.
Generally, as I understand it, it takes about 5-7 times longer to test (on the basis of 10 pairs).
Why does everyone strive to make multicurrencies, is there any substantiated evidence of their effectiveness or...?
1. Not all, most do and will do monovolutions.
2. In the light of the new features of MT5 mults will eventually get more chances than before (a normal mult in MT4 is almost no real task), but due to complexity of implementation (and its minimally higher by an order) many of the experienced EA writers will not undertake such projects.
3. There are also differences in hedging and trading opportunities (in the championship a monovaluator is limited to 15 lots, while a multicurrency one easily overcomes this limitation).
Answering all at once.
Thenumber of orders is optimised - now I have even disabled trailing, so only entry/exit.
Multicurrency is imho more reliable than single currency, because you get a portfolio of strategies with different settings - if a couple of instruments fail, the others can help. But that is not the subject of discussion here.
This is how the initializer becomes a bottleneck - I want to know more about it. I use one indicator, but it's read from it in those 1.5 minutes of timing that I measured.
- www.mql5.com
I have one indicator in use, but the reading from it is included in the 1.5 minute timings I have measured.
A multicurrency is imho more reliable than a single currency, because you get a portfolio of strategies with different settings - if a couple of instruments "fail", the others can back it up.
That's what I think too, but is that a fact?)
1. Not all, most do and will do monovolutions.
2. In the light of the new features of MT5 mults will eventually get more chances than before (a normal mult in MT4 is almost no real task), but due to complexity of implementation (and its minimally higher by an order) many of the experienced EA writers will not undertake such projects.
3. There are also differences in hedging and trading opportunities (in the Championship the monovaluator is limited to 15 lots, while the multicurrency one easily overcomes this limitation).
This is how the inidactor becomes a bottleneck - I'd like to hear more about this. I use one indicator, but the data reading from it is included in those 1.5 minutes of timing that I have measured.
The indicator should be included directly in the Expert Advisor code and use only the last bars in the calculation. The performance of indicators on the example of MA is studied here.
2. You should not perform unnecessary calculations on every tick. You do not place or close orders on every tick, do you?
The multicurrency strategy was present in all the championships and where is its extra profitability?
1. The same is true for the EA, but the EA's performance and reliability compared to, say, 10 EAs not sharpened for portfolio work, can be argued for.
It is true, as noted above, this is not the topic of this branch.
2. Seriously about the role/weight of multicurrency in the championships can be discussed only from Championship 2010 onwards (because it was the first championship that officially allowed multicurrency strategy testing).
But even Ch2010 will not be representative from this point of view.
IMHO
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
This is an interesting situation. My multicurrency Expert Advisor, unencumbered by any online retraining, does not fit into the 15 minute limit.
I inserted a primitive profiler and surrounded my code calls with it. It turns out that my code works for only 1.5 minutes for a 16-minute test from the beginning of the year. Where does the rest of the time go? And how can I optimise something which is not in EA?
Do you have any ideas?