How was your first EA developed? - page 6
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
Although I've only dabbled in Python, my understanding (based on the wealth of relevant information on mql5.com) is that Python is king for implementing highly complex trading systems for use with MT5.
I also understand that there is some backend Python support built into MT5, but I've never used it.
For an average trading system, straight MQL5 is the way to go.
I've also noticed that most of the articles that implement neural networks or machine learning connect Python to MT5.
Yeah, Python has a dedicated library for MT5. Besides connecting both, it can also trade, extract data, and do a lot of things through Python.
Once you have the data, you can analyze it using all the Python stuff.
The part most difficult for me was battling with overfitting.
Overfitting is common among machine learning EA's, and over-optimizing is even more common among all EA's.
I really don't even optimize much at all. Based on my experience, I can generally deduce whether experimental logic can make it nor not based on a single backtest.
Spoiler alert: Profitable EA's are just as rare as profitable traders. At least, parts from the EA salvage yard can be useful in the future.
Overfitting is common among machine learning EA's, and over-optimizing is even more common among all EA's.
I really don't even optimize much at all. Based on my experience, I can generally deduce whether experimental logic can make it nor not based on a single backtest.
Spoiler alert: Profitable EA's are just as rare as profitable traders. At least, parts from the EA salvage yard can be useful in the future.
What do you think is the main thing that separates profitable EAs from unprofitable ones?
I suspect that what you're really asking about is... the main thing that separates profitable trading systems from unprofitable trading systems.
Let's back up a bit and first consider the: (1) market, (2) instrument, and (3) chart timeframe or custom chart structure to be traded.
I know that I'm in the minority when I say this but, every EA should be specifically designed specifically for those 3 (4) things. If it's not, good luck.
Edit: (4) specific broker-dealer─for OTC markets.I suspect that what you're really asking about is... the main thing that separates profitable trading systems from unprofitable trading systems.
Let's back up a bit and first consider the: (1) market, (2) instrument, and (3) chart timeframe or custom chart structure to be traded.
I know that I'm in the minority when I say this but, every EA should be specifically designed specifically for those 3 (4) things. If it's not, good luck.
Edit: (4) specific broker-dealer─for OTC markets.Do you consider good making an strategy for a pair and then adapting it for other pairs using the strategy tester?
I have the idea of making a portfolio of the same strategy but in differente pairs.
Do you consider good making an strategy for a pair and then adapting it for other pairs using the strategy tester?
I have the idea of making a portfolio of the same strategy but in differente pairs.
I consider that a hail Mary when my intended FX pair shows little promise─basically, an act of desparation.
TBH, I try to avoid it.
I consider that a hail Mary when my intended FX pair shows little promise─basically, an act of desparation.
TBH, I try to avoid it.
What about using different strategies on the same pair?
I can't really imagine what the point of that would be... for me. I'm in a Netting-only jurisdiction─all positions in currency-sharing pairs offset each other.
Even in a Hedge account, multiple strategies could be implemented within one EA. To the extent that they would still remain multiple strategies, ok I guess.
I can't really imagine what the point of that would be... for me. I'm in a Netting-only jurisdiction─all positions in currency-sharing pairs offset each other.
Even in a Hedge account, multiple strategies could be implemented within one EA. To the extent that they would still remain multiple strategies, ok I guess.
Ok, ok.
I'm asking this more in the sense of diversifying in order to reduce risk. I have seen that some people use many pairs with the same strategy, only modifying the parameters, while others use different strategies with negative correlation on the same pair. However, I haven't tested that approach very much.