"The 'perfect' trading system - page 29
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
The "tactile sensation" is only in the story.
The driver looks ahead with his eyes - he can see a piece of the future road.
The trader cannot see a piece of the future price range - there are no prices to the right of the terminal window.
So you can talk about what you "feel" there (to the right of the terminal window) somewhere on RBC, for example, in the Investor's ABC :)
Very authoritative conclusion from a man who feels the future by tactile sensations with TA :)
TA in general only increases misalignment, i.e. it increases losses, which is what statistics show us - people believe you can predict the future or feel it with your fingertips there on the right-hand side of the monitor :)
However, what does not yet exist cannot be felt, either tactilely or non-tactilely.
NF is a function that you create yourself, i.e. you trade - you get a numerical series.
For example, "buy cheaper, sell dearer" - the result of this strategy is NF.
The trajectory of your car is also SF.
Any result of traders' trading in contests is their SF.
You can use the PRNG to create SF, for example as follows:
1. set the direction of the PRNG trade, if more than 0.5 - 1(buy), less than 0.5(sell)
2. Limit and stop are equal to the constant.
The result will be some random curve, which in general will rarely match the FF, i.e. according to statistics 95% of traders lose.
In fact, these 95% of traders are just using their own PRNG, which they loudly call a "trading system".
The main NF in the EA is a sine wave. StopBase is 1/4 wavelength.
The sine wave is synchronized by the FR, therefore it always has different final wavelengths and shape - it is compressed, then uncompressed, respectively the amplitude is less, then more.
I'm too lazy to draw, so I'll have to "strain" my imagination.
Any periodic function will do instead of a sine wave - it's easier to implement a periodic function, because it requires fewer parameters - a sine wave requires only one.
Comrade, I'll tell you this - if you have no special education in the field of radio electronics (and judging by your posts you either don't have it or have a very mediocre one), do not try to improve yourself in the eyes of others by using special terms that you read in the encyclopedia. Trying to describe the frequency-phase automatic tuning, you do not understand it yourself, and therefore you start to weave in here clever names of entities which you have invented yourself. You'd better write it like this: not an expert, I explain everything, "in my own words". Otherwise, you sound like a psychic who uses words like "energy", "information", etc., without making any sense except for his vague sensations, and fills the brains of listeners with IQ <100. You've got the wrong place; most of the people in the discussion are people with higher technical education and a lot of experience. So by saying that your system (whatever it is) gives a positive result and yet does not approximate the market, you are demonstrating only one thing - once again - you don't understand what you are doing.
My free advice to you: explain on your fingers what you want to do, knowledgeable people will tell you what it is called "scientifically" and maybe even help you practically.
Read it more carefully.
It was in the context of extrapolation ("FR synchronises SF, without extrapolation").
About BSE. You needed a reference to the term - the search engine gives references mainly to BSE or wikipedia.
Spending time to find references to special literature is not very desirable to me - that you somehow look for yourself.
Why are you arguing with him?
:-)
I am not arguing, but sincerely trying to understand what the man has behind him. Judging by the above-mentioned website, VictorArt is Victor Artyukhov, and he has been developing his digital brain and systems based on it since 1993. (Well, if I'm wrong, we're debating an impostor.)
He has posted the MTS Constructor here and links to PAMMs. Enough to know that there are some performance results.
One could of course try to deal with those results. However, my mind is more interested in theory. If the theory is unimpressive, then there is no confidence in the practical results. In any case, even though they are clearly positive, they cannot be explained by the proposed theory.
So I want to understand what it's really like.
:-)
I am not arguing, but sincerely trying to understand what the man has behind his soul. Judging by the mentioned website, VictorArt is Victor Artyukhov, and he has been developing his digital brain and systems built on it since 1993. (Well, if I'm wrong, we're debating an impostor.)
He has posted the MTS Constructor here and links to PAMMs. Enough to know that there are some performance results.
One could of course try to deal with those results. However, my mind is more interested in theory. If the theory is unimpressive, then there is no confidence in the practical results. Anyway, even if they are clearly positive, they are in no way justified by the proposed theory.
So I want to understand how it is in reality.
Historically, it was like this:
1. the MTS constructor appeared, using CM + classic TA - I knew nothing about trading at all then, so everything was done based on the knowledge of consultants - professional traders with a long experience
2. the results were good, but all the time some strange - then a lot of profits, then large drawdowns.
3. they started to dig
4. the General Theory of Trading (GTT) was introduced and some not very clear results, but very good ones
5. then everything was done only within the framework of the oTT
6. it is now possible to automatically create new adaptive trading robots
7. And then, it took me less than 3 hours to write the adaptive EA code and it immediately started working as expected - even a little better.
So, personally, I'm quite impressed with OTT as we can use it to get almost useful predictable results and accordingly, use it and get better results every day.
Genius! :)
The only thing left to figure out is where the trend will be and how long it will last...
But TA will no doubt help you with that :)
It's very simple.
It's as simple as that.
If it's simple, then what's the problem - trade only trends - because you know their beginning and end.
However, because 95% don't want to - they lose :)
NF is a function that you create yourself, i.e. you trade - you get a numerical series.
For example, "bought cheaper, sold higher" - the result of this strategy will be SF.
Any result of traders trading in contests is their SF.
...
The basic SF in an EA is a sine wave. StopBase is 1/4 wavelength.
The sine wave is synchronized by the FR, therefore it has different final wavelengths and shape all the time - it is compressed, then uncompressed, respectively the amplitude is less, then more.
Any periodic function will do instead of a sine wave - it is easier to implement a periodic function, because it requires fewer parameters - for a sine wave one is enough.
Well, this is more specific.
As I understand in the first sentence of the post - this is just an example of SF after all. One of them, so to speak. As for your EAs, the SF there is a sine wave. I draw my attention to it because if the trade result is a sinusoid, the profit of such a trade is zero. It means that these two NFs have different functions. And if we talk about the trading result, the desired SF there is an ascending straight line.
I have no objections to a sinusoid as the SF, it is the first indication of the oscillatory character of the price movement. So as a model (or NF, if you prefer) it is quite a suitable function.
From this follows an answer to a question that has been of interest to me since the beginning. By its very name MTS Constructor, implements a user-defined MTS. Naturally, this MTS has its own SF. The question was, can the user of the MTS Builder use it to put any desired SF into the MTS being built? Or is this NF predefined as a sine wave after all ? From your post I understand that it is predetermined.
If not, please explain how it is possible to use Constructor to put your NF into MTS. Looking through the Builder documentation I didn't see such a possibility. Only setting values of parameters provided there.
We have SF - a sine wave.
FR synchronizes SF.
If the SF is also a sine wave, it is quite easy to synchronize them.
In our case, the SF is an unknown shape in advance, so the SF synchronizes the SF in such a way that the SF is not very far away from the SF.
In the simplest case, a stop loss triggering is enough - we change direction and for a while the NF is again "inside" the FR...
...
Periods of synchronism are periods of profit.
Periods of asynchrony are periods of losses (car behind a fence, somewhere in the woods).
Optimization in this process is only needed to minimize the size of the stop loss, i.e. reduce the cost of synchronization - the more accurate the synchronization (less cost) - the greater the profit.
If you read carefully, you should immediately realize that during the flat, the synchronization costs are greater than during the trend. Hence the expression: "trend is a trend".
Yes, it's all clear.
As far as I understand, a stop loss triggering, profit and loss chains are the same signals or criteria used for synchronization. I.e., synchronization is a very expensive process and, by the way, it must continue all the time - adaptation cannot stop. Therefore, the questions are. Does your system have other ways of doing synchronisation during trading, apart from the losses you receive? What methods do you use to minimise the loss and maximise the profit of each trade ? It is not about the aggregate result, but about making the average profit on profitable trades higher than the average loss on unprofitable ones. You seem to be OK with the profit factor, but there is also this side of MM.