Dependency statistics in quotes (information theory, correlation and other feature selection methods) - page 45
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
I have never understood such conclusions, "If it were easy - everyone was a millionaire, the funds would have calculated everything long ago, they have a lot of analysts and power and mathematics and so on and so forth...".
First of all you won't get that kind of money in the kitchens, and then why take the technology of earning money from us singles and so casually equate it with the technology of funds trading.
Take futures, do you think with equally profitable signals you can realise any amount in a trade, who told you that, liquidity has limits. On the one hand there is a liquidity constraint, on the other hand there is a window of graality. If we have an advantage, but the time of this advantage may be less than the time to deposit the necessary amount of money (not necessarily instantaneously, but in series) to get a plus in the end of the series, here you have the limit, but for us this advantage is small enough to live normally, and the funds do not really care. The games are completely different there.
The man has already shown https://www.mql5.com/ru/forum/139328/page7 how he had reached the liquidity limit, and this, mind you, even on such the most highly liquid instruments in the WORLD.
1. to obtain a trading system in which the order will be at breakeven, say, 7 out of 10 times is not as difficult as it seems, in such TC balance chart has the form of a series of rectangular triangles (earned, earned, earned, drained all that earned, earned, ..... ). The minimum target is the minimum risk. It is better to consider real "task minimum", clearly defined profit/loss ratio ---> takeprofit/stop loss, otherwise another study_of_the_spherical_horse_in_the_vacuum
2. profit is good, but I have never met a FC, in which there is no series of losses, only profitable, the so-called "Grails" (I know, I have seen the grailers, but they do not have series of losses, but immediately lose their deposits ))))). ). My point is that the maximum objective is to reduce the number of consecutive losing trades.
Everything is not simple at all, if it were simple - the Russian Pension Fund would earn on the markets, instead of losing ((
Let's talk about the topicstarter's first post. Did you find any incurable errors there?
Firstly, it's price returns that are being investigated, not the price itself. That's the object of the study. Looking for further arguments to show that price cannot be investigated in the same way.
there is a science of "formal languages" which studies the formation of different languages and their interpretations.
Any language makes sense when there is a strict agreement on the rules that are followed by native speakers. If this is not the case, the only thing the new alphabet can be used for is data flow compression, such as archivers. I.e. eliminate redundancy.
Otherwise, we just jump from one data representation to another, without extracting any additional information.
The task of a language is to describe real processes. To do this, one native speaker has to describe them, and the other, once received, understands them. This is not the case in the marketplace. The processes are there, but there is no common language and no description of the processes by the native speaker. Therefore grammars are useless for this
How do you rate the performance of EAs in the search box, type sulton The results of 58 EAs will appear. So far most of the results are over 50%. I think your idea that the maximum objective is to reduce the number of series of continuously losing trades has almost been realised.
yosuf : Are you now actively testing your 58 EAs for moderation? ;)
Mathemat: " Firstly, it is the returns on the price that are being studied, not the price itself. This is the object of the study. I'm looking for additional arguments to show that price can't be investigated in the same way."
It's a matter of taste, of course. Someone is looking for ways to analyze the price, someone is looking for arguments why it cannot be analyzed. Everything has a right to life, even when proven otherwise.
But if "price also cannot be investigated", how can the results be applied to it?
And the results themselves are not clear how to interpret them.
It's a matter of taste, of course. Some are looking for ways to analyse the price, some are looking for arguments as to why it cannot be analysed. Everything has a right to life, even when proven otherwise.
But if "the price also cannot be investigated", how can the results be applied to it?
Firstly, Mathemat has said that he is looking for further arguments, which means that his point will be valid in any case. And secondly, we (me, if you like) investigated a stationary process consisting of price increments. The results of the interpretation are clear as day (you just haven't figured it all out yet). If the model predicts the next letter of the alphabet corresponding to an up move, enter long, if the letter corresponds to a down move, enter short, if the letter corresponds to a small move within 2 spreads, do not enter.
to Avals:
And I still think any trader can understand what it means to assign the value of the price increment between adjacent opening prices to one of several levels: from 1 (strong move down) to e.g. 7 (strong move up), where 4 would be a move within a spread-two.
to VNG:
Thanks for trying to make sense of what I was doing. But, examining the original price values is dangerous in that it is not a stationary process, and if you have given a price of 0.954 an alphabetical index of 154, you may find that this "letter" will not be used by this very process for the next 8-9 years. Oops. And our approach was based on the fact that the form of the probability distribution function on the alphabet letter space was - initially - close in shape to the distribution of price increments in Forex, and then - became uniform. That is, all the letters were used.
yosuf : Are you now actively testing your 58 EAs for moderation? ;)
Mathemat: " Firstly, it is the returns on the price that are being studied, not the price itself. This is the object of the study. I'm looking for additional arguments to show that price can't be investigated in the same way."
It's a matter of taste, of course. Someone is looking for ways to analyze the price, someone is looking for arguments why it cannot be analyzed. Everything has a right to life, even when proven otherwise.
But if "price also cannot be investigated", how can the results be applied to it?
And the results themselves are not clear how to interpret them.
They are tested for unpretentiousness (all tools are used) and reliability compared to other Zulu EAs (there are about 15000). In a month or so mine have outperformed most of them in the hard Zulu ranking.
It's not simple at all, if it were, the RF Pension Fund would be making money on the markets, not dumping it (((
What does that have to do with the RF Pension Fund?! And what makes you think that even if things were simple, the PF would make money?
Grails, pension funds... Soros is also forgotten ;)
They are tested for unpretentiousness (all tools are used) and reliability compared to other Zulu EAs (there are about 15000 of them). In a month or so mine have outperformed most of them by hard Zulu ratings.
What does this have to do with the issues raised in this thread?
A medal on your chest, a band to help you and to the platform to meet the trains.
Firstly, Mathemat has said that he is looking for further arguments, which means that his point will be valid in any case. And secondly, we (me, if you like) investigated a stationary process consisting of price increments. The results of the interpretation are clear as day (you just haven't figured it all out yet). If the model predicts the next letter of the alphabet corresponding to an upward move, enter long, if the letter corresponds to a downward move, enter short, if the letter corresponds to a small move within 2 spreads, do not enter.
Alexey, Mathema and I have "peace, friendship" too;).
The thing is, if your results were to give a qualitative prediction, then I would study your approach. But so far this has not happened as you can see from the results. So you and I are talking about the approach as a whole to understand - methodological error, hypothesis, execution, or whatever else did not give the expected result. Do I understand the purpose of the discussion correctly?
"... to investigate the initial values of prices is dangerous because it is not a stationary process..."
What is the danger?
There is no need to be afraid to work with price;)