What to feed to the input of the neural network? Your ideas... - page 20
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
to optimise the algorithm - only fixed volumes, and no averaging or locks. otherwise you will get this and this.
Looks like it's fixed lots. He did x4, so the risk is reduced. But it should be in % of deposit - then the risk will always be equal, for example 5% of deposit.
The realisation seemed strange. I couldn't get through the articles at all, as if the author was speaking a higher language. But, the description of the technology: "a neural network with many outputs and each responsible for a specific action."
Where are these actions? Why is there no action to buy 0.6 lots? And not 0.01. Why there is no action "Close part of position, leave 0.07 lot in the market", why there is no action "open an additional position, risk is justified".
Dynamic lot, dynamic TP-SL (or reverse signal), position maintenance - there is nothing.
Instead - buy-sell. Everything.
Although, recently there was an article "Moral Expectation at Forex". Even there they tried to explain somehow mathematically the usefulness of dynamic lot and dynamic TP-SL. After all, a trader does all this.
Therefore, I would like to see this implementation - to make sure that everything does not work to hell. I'm going to study matstat and the essence of MO.
The variable lot is secondary and will not fundamentally change anything.
Two situations:
1. A month in the minus almost, the last three days, again similar signal to open, but increased lot - did not go long, a few points in +, closed on the reverse signal: closed the month in +, pulled out the deposit
2. A month in the minus almost, the last three days, again similar signal to open, but a regular lot - did not go long, a few points, closed on the reverse signal: a transaction kopecks in +, but the total minus did not cover, did not pull out the deposit.
And if during the month there will be a variable lot - it is a completely different picture of the balance will come out with the corresponding reaction of the advisor to the drawdown or growth.
If you become a successful trader, you will have clarity), but for now it is an open question.
It's a tester. And in a tester you can show any beauty you want.
I don't argue, yesterday I myself created beauty ON REAL TICKETS in the strategy tester.
I thought for a long time how it is possible. It turned out to be simple - you just need to try the same mode on real ticks... on a real MT account, not on a demo)))
I understood that commissions do not allow tick scalpers to work on impulses. But I didn't realise that MT5 has different conditions for tester on demo and tester on real account.
The easiest way to feed digitised bars to the neural network input. Simple and convenient.
Please explain how it differs from the first post of the thread?
I mean that I tried closing prices and converted closing prices (-1.0..1.0 or 0.0..1.0).
Could you please explain how this differs from the first post of the thread?
I mean I tried closing prices and converted closing prices (-1.0..1.0 or 0.0..1.0).
That's not it. You have to create numeric names for different bars and then analyse them. This is a hint.
I will not reveal the highlight. Because some of the shots here tried very hard to spoil my blood.)))
We need to create numeric names for the different bars....
Exotic.
I was going to write "original", but it doesn't quite express surprise.
I have no idea what that is. A digital signature of a person's identity comes to my mind, but how does that go with all the bars...
I won't reveal the highlight. As some shots here tried very hard to spoil my blood)).
Exotic.
I was going to write "original", but it doesn't quite express surprise.
I have no idea what this is. I can think of a digital signature of an individual, but how does that go with all the bars....
Then maybe I'll knock in personTomorrow. I'm going on holiday.
Guys decided to invent clustering.)
Maybe it's got a gimmick, clustering 2.0.
I'll find out tomorrow.