Discussing the article: "Evolutionary trading algorithm with reinforcement learning and extinction of feeble individuals (ETARE)"

 

Check out the new article: Evolutionary trading algorithm with reinforcement learning and extinction of feeble individuals (ETARE).

In this article, I introduce an innovative trading algorithm that combines evolutionary algorithms with deep reinforcement learning for Forex trading. The algorithm uses the mechanism of extinction of inefficient individuals to optimize the trading strategy.

I remember that day like it was yesterday: June 23, 2016, the Brexit referendum. My algorithm, based on classic technical analysis patterns, confidently held a long position on GBP. "All the polls show that Britain will remain in the EU," I thought then. At 4 a.m. Moscow time, when the first results showed a victory for Brexit supporters, GBP collapsed by 1800 points in a matter of minutes. My deposit lost 40%.

In March 2023, I started developing ETARE - Evolutionary Trading Algorithm with Reinforcement and Extinction (Elimination). Why elimination? Because in nature, the strongest survive. So why not apply this principle to trading strategies?

Are you ready to dive into the world where classic technical analysis meets the latest advances in artificial intelligence? Where every trading strategy struggles for survival in Darwinian natural selection? Then fasten your seat belts – it is going to be interesting. Because what you are about to see is not just another trading robot. It is the result of 15 years of trial and error, thousands of hours of programming and, frankly, a few destroyed deposits. But the main thing is that it is a working system that already brings real profit to its users.


Author: Yevgeniy Koshtenko

 

Good evening.

I read your article and found it extremely interesting.

I have set up the consultant to work on a demo account. It opens trades correctly, but does not close any trades, regardless of the balance, whether it is positive by a few thousand euros or negative. I would like to know if it works like that and if the trades should be closed manually, or if I need to change some parameter in the Python file and set the profit or stop loss, either globally or individually.

I don't speak Russian, only Spanish.

This message has been translated by Google, but I hope it is understood. Thank you very much for your attention and dedication.

 

This approach and methodology are simply brilliant. Their implementation will accelerate the self-learning and modernisation of the system, making it ever more advanced. It really seems to be able to outperform the market. Great work! I am not yet familiar with using this EA and database. Could you please provide me with the full workflow and EA for testing? Thank you very much.

 
An intriguing approach, thanks to the author for his contribution. However, the code is just a Python class, unusable without an EA and a DBMS. I hope, in the future, the author will provide us with a working system or at least some guidance for implementing and experimenting with his evolutionary approach. Thanks in any case.
 

Hello greetings from Indonesia,

I was look your algorithm and look like seems great article.

May i got ur github link ? thanks in advance 

 
Hello, can you provide the MetaTrader5 package for python please?
 
xiaomaozai #:
Hi, can you provide the MetaTrader5 package for python please?
https://www.mql5.com/en/docs/python_metatrader5
Documentación para MQL5: MetaTrader para Python
Documentación para MQL5: MetaTrader para Python
  • www.mql5.com
MQL5 ha sido pensado para el desarrollo de aplicaciones comerciales de alto rendimiento en los mercados financieros y no tiene análogos...
 
Applying evolutionary theory to strategy writing, extinguishing mistakes, and playing to strengths, what level of evolution will ultimately take place? Looking forward to it.