NATURAL INTELLIGENCE as the basis of a trading system - page 92
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
We have extracted one possible answer from Papa_Jozh's link above:
S.Lyubimov:
"
If a trader has a desire to learn, they must first and foremost learn. But in order to do so, one must initially have a great deal of common sense and creativity. In my view, a formal education in economics is of little use here, although general erudition and familiarity with the mathematical apparatus never hurt.
You have to realise from the outset that the market is a very competitive and noisy environment and try to find your own qualities which can give you market advantages. You have to develop them. Or, immediately find a job in the market on the other side of the counter - say, in a brokerage firm.
Then, at the expense of the employer you can learn something and, if there is strength and desire,
to try and make a career for yourself.
"
== i.e. the time frame for achieving insight is comparable to that of a career, == calendar years in which academic qualifications are not included.
== i.e. the time frame for achieving insight is comparable to that of a career, == calendar years which academic qualifications are not included.
I don't think so.
I think the reference to common sense and creativity is worth paying attention to.
I don't think so.
I think we should pay attention to the reference to common sense and creativity.
.... and on natural selection in the trade process)))
The basic learning algorithm is repetition.
Bravo!
It is the repetition of a sequence of events, among which there are biologically relevant limiting events, that allows the organism to isolate signals from the external (or internal) world which warn the organism of danger long before it realises it. The organism predicts danger.
Peter Kuzmich Anokhin called it "anticipatory reflection".
"Essays on the Physiology of Functional Systems."
The principle of anticipatory reflection underlies all information processes in the organism.
And functional system is a universal architecture, which implements this process at all levels of organization, including the neuron.
Existing models of neurons used in neural networks lack fundamental elements and mechanisms of the functional system.
Therefore, neural networks remain just a mathematical toy.
I am using functional system modelling in my trading systems. And all learning is based on 200-500 days of tick-data repetition.
The systems attempt to extract signals to execute trades. The result of the stock is associated in the memory apparatus with the signal that triggered it.
In the real market, the systems try to recognise familiar situations and execute the trades.
The behaviour of my systems allows me to call them intelligent.
I hope I have answered some important questions of this thread.
Thank you for your attention.
....
Thank you for your attention
Our attention is at your service.
How is the progress of your sensible systems?
Our attention is at your service.
How are your sensible systems progressing?
it's here:
http://gointernational.com/SctRes/default.htm
After training.
In reality, it's 4-5 times worse.
it's here:
http://gointernational.com/SctRes/default.htm
After training.
In reality it is 4-5 times worse
I can't say that the tables in the link are clear to me, but one more clarifying question: what does "4-5 times worse in real life" mean - do you work in real life with this system?
I can't say that the tables in the link are clear to me, but one more clarifying question: what does "In real life it is 4-5 times worse" mean - do you work in real life with this system?
Yes, I've been doing it for a long time :)
And the tables show the results for each day, with full details of all stocks by minutes and total for the entire period
It is easy to check with the historical data.
Our attention is at your service.
How is the progress of your sensible systems?
By the way, dear rsi (just noticed), have you tried the RSI(3)-RSI(1) indicator?
It's quite interesting - sometimes it allows to predict the peaks (no kidding).
No, I don't use such an indicator.
Still, I don't understand where your bot uses the "200-500 days tick-data" information. I think it's just pipsing. What broker tolerates you?