From theory to practice - page 313
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
It turns out we need to work on the MM and allow simultaneous trades.
There's not much point in allowing them that way, since the result will also be correlated.
No, opening with the same lot on 4 correlated pairs gives less drawdowns and rollbacks in equity than opening with a quadruple lot on one of the pairs.
Although they are correlated, they often have a lag relative to each other.
toAlexander_K2
But in the context of this MM branch it is a principal step aside (this is a message to Alexander).
(That's a message for Alexander.) Cut them, they're gold )))
Not so, opening with the same lot on 4 correlated pairs gives less drawdown in the moment
Yes, that's right, I meant that one should not wait for great effect, diversification is there, but weak.
I agree, MM is a grind of the ready-made TS.
Much more important is to build the TS itself. My criteria: TS is ready if on forward, constant lot, on one pair, the shape of growing balance chart, does not change character, compared with the optimization section. Then I can add more tricks.
Herehttps://www.mql5.com/ru/forum/86386/page836#comment_7062634 I suggested experimenting with Erlang flows. Is this the way you mean?
Thanks, I'll have a closer look, I understand that@Maxim Dmitrievsky has started to work in this direction, and are there any results? I think it is possible only when the resulting distribution is close enough to Gauss.
As for your system, I meant the following: You drive an exponent through a sample, where part of it is logarithmically distributed, and part where the "tail" itself there are single outliers of large delay amplitude, not synchronous, asymmetrical, and there you also drive an exponent, but this process is already different (different from non-random), different from the main sample, and this exponent is simply stretched on them, I do not know how else to put it better, I hope you will understand me.
Thanks, I will have a closer look, I understood that@Maxim Dmitrievsky started to work in this direction, and are there any results? I think it is possible only when the resulting distribution is close enough to Gaussian.
As for your system, I meant the following: You drive an exponent on sample where a part is logarithmically distributed, and a part where "tail" itself there are single outliers of the large delay amplitude, not synchronous, asymmetrical, and there you also drive an exponent, and this process already other (different from non-random), different from the basic sample, and this exponent is simply stretched on them, I do not know how else better to put, I hope you will understand me.
Haven't started yet. I need a proper transformation formula for indicators, csv files are not convenient to work with
They are probably presented here in the thread, but I have not looked for them yet )
For statistical enthusiasts I am attaching the converted tick BP to Erlang flows of various orders.
In the file name the first two digits are the order of Erlang flow.
For example: 01 AUDCAD ... - simple flow, 30 AUDCAD ... - 30th order Erlang flow
Data processed for 02-06 April 2018.
You can observe how the distribution of increments changes as the order of the flow increases. I recommend it. Very interesting.
If necessary, I can continue flow transforms up to 100th orderFor statistical enthusiasts I am attaching the converted tick BP to Erlang flows of various orders.
In the file name the first two digits are the order of Erlang flow.
For example: 01 AUDCAD ... - simple flow, 30 AUDCAD ... - 30th order Erlang flow
Data processed for 02-06 April 2018.
You can observe how the distribution of increments changes as the order of the flow increases. I recommend it. Very interesting.
If necessary, I can go on with flow transformations up to 100th order.Alexander.
Actually, first create a robot, then test it, look at equity and balance graph, and only then draw conclusions.
Your approach - belief in super genius - is not supported by anything.
Usually this approach carries a decent amount of unreasonable monetary costs!Alexander.
Actually, they first create a robot, then they test it, look at the equity and balance graph, and only then draw conclusions.
Your approach - belief in super genius is unsupported by anything.
This approach usually carries with it a decent amount of unreasonable monetary costs!I suddenly have super-genius powers from my thirst for money. Rena, get your pockets ready!!! Both my eyes are already twitching with anticipation of easy profits.
My thirst for money has suddenly given me super-genius powers. Rena, get your pockets ready!!! Both my eyes are already twitching with anticipation of easy profits.
I've already seen your little signal...
The thirst is there, but the money is less and less.
Take off the rose-coloured glasses, it's about time.