Discussion of article "Applying the Monte Carlo method for optimizing trading strategies" - page 2
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
Alexei, thank you for the material. There is a lot to think about.
I was impressed by the conclusion in the "Conclusion" block:
Моделирование методом Монте-Карло было бы более полным, если бы имелась возможность подавать на вход советника не только имеющуюся историю цен, но и её копии, изменённые случайным образом. Это позволило бы изучать устойчивость советников более основательно, хотя и ценой увеличения объёма вычислений.
Now that's a really interesting challenge! And I think it is solvable. I assume that it can be solved with the help of custom characters.
Now that's a really interesting challenge!
MonteKarlath has been doing exactly that since the time of Gorokh. The Expert Advisor is run on a bunch of generated symbols with similar statistical characteristics.
A delusional method, but it's good for books and seminars.
MonteKarlath has been doing exactly that since the time of Gorokh. The EA is run on a bunch of generated symbols with similar statistical characteristics.
It's a delusional method, but it's good for books and seminars.
Monte Carlo simulations are especially inapplicable in the case of data-mined strategies and multiple comparisons. Actually, if these simulations are used with data-mined strategies, it is a strong indication that the developer lacks experience. In a nutshell, over-fitted strategies usually generate good Monte Carlo results. When this method is used in a loop of multiple comparisons, it loses its significance completely.
https://towardsdatascience.com/validation-methods-for-trading-strategy-development-1efea8284b02
Dennis Kirichenko:
I assume it can be solved with custom characters.
You are right, that's what I was thinking about when I wrote this. I didn't write about it because I'm not ready to touch on this very extensive topic yet.
Monte Carlo simulations are especially inapplicable in the case of data-mined strategies and multiple comparisons. Actually, if these simulations are used with data-mined strategies, it is a strong indication that the developer lacks experience. In a nutshell, over-fitted strategies usually generate good Monte Carlo results. When this method is used in a loop of multiple comparisons, it loses its significance completely.
https://towardsdatascience.com/validation-methods-for-trading-strategy-development-1efea8284b02
Yeah, Monte for Carlo.
https://medium.com/@mikeharrisNY/fooled-by-monte-carlo-simulation-eee0b312e489
It's a delusional method, but it's good for books and seminars.
There is some truth in what you say. But they say more about the writers of books than about the method. I am sure that if you take a sensible approach to designing a method of price generation, you can benefit from the method.
Monte Carlo simulations are especially inapplicable in the case of data-mined strategies and multiple comparisons. Actually, if these simulations are used with data-mined strategies, it is a strong indication that the developer lacks experience. In a nutshell, over-fitted strategies usually generate good Monte Carlo results. When this method is used in a loop of multiple comparisons, it loses its significance completely.
https://towardsdatascience.com/validation-methods-for-trading-strategy-development-1efea8284b02
Quite a good advert for the book. But I would like to see concrete arguments and, most importantly, what is proposed instead of Monte Carlo.
There is some truth in what you say. But they say more about the book writers than about the method. I am sure that if you take a sensible approach to designing a method of price generation, you can benefit from the method.
As well as porridge from an axe can turn out great.
Quite a good advert for the book. But I'd like to see the specific arguments and, more importantly, what is proposed instead of Monte Carlo.
Books? I think there are only articles
fxsaber gave the arguments in his other article.
I don't know about other methods. if I did, I would have already attached them :)
As well as porridge from an axe can turn out great.
At the moment I am not ready to give you a substantive answer.
In the article you cited, the author does not come to the conclusion that it is necessary to abandon the Monte Carlo method and replace it with some other one. He only speaks about the necessity to be more careful. In principle, the idea is correct though not new.