Dependency statistics in quotes (information theory, correlation and other feature selection methods) - page 54

 
VNG:

Tadhv, Vadimchi's channels and swings are in the public domain, but they haven't stopped working and never will. They are eternal. Because they have an indestructible foundation at their base and are built on patterns that are outside the market.

Yep, a perpetual motion machine. Give this system to all market participants and there will be communism :)
 
Avals:

In order to make a profit, there has to be a directional move. That is, this pattern signals a time when quite a few people will start buying or selling on the market. What could possibly encourage them to do this regularly, in any market, and forever)?

Macroeconomics, lust for profit + hype on the news.

Back to basics))))

 
VNG: I forgot, decided to add to it. It is for forecasting that I want to apply TI to patterns.

I would like to do it too. But first I would like to learn how to predict at least one bar.

Then you can get down to a few: memory is extremely long-term, with thick tails; it doesn't matter which bar to predict, zero or minus fifty.

 
sergeyas:

Macroeconomics, lust for profit + hype on the news.

Back to basics))))


So markets are so different in the composition of the participants, the way they make money, the impact of the macroeconomy, the microstructure of the market, that universality is utopian. And so is the eternity of work, since all this changes over time
 
Mathemat:

I'd like that too. But first I would like to learn how to predict at least one bar.

Then you can get down to a few: memory is extremely long-term, with thick tails; it doesn't matter which bar to predict - zero or minus fifty.

The Emergency has been renamed - it's called FARIMA. That's where the tails are, well, very thick.
 
alexeymosc:

Good afternoon!

I decided to slightly develop the topic touched upon by Alexey (Mathemat) in one of the forum threads.

I tried to search for dependencies in the quotes of one financial instrument using statistical methods. To begin with, I took the Dow Jones Industrial index, daily data, and transformed a series of series into the series of percentage increments.

The article is actually here: http: //habrahabr.ru/blogs/data_mining/127394/

I would like to continue for FX quotes, I will post the results here.

I have viewed the article referred to by the topicstarter.

I can't help thinking that it's just a numbers game. The article states a trivial idea - the increments in the quotient are dependent. The calculation of the dependence is made by some formula, and there is no justification that it can be applied. The most important rule of statistics is violated: any result must have a meaningful interpretation. And what is it here?

 
VNG:

Tadhv, Vadimchi's channels and swings are in the public domain, but they haven't stopped working and never will. They are eternal. Because they have an indestructible foundation at their base and are built on patterns that are outside the market.
Where can I see the results of their work? Is it on onyx, for example?
 

sever32:
what is it? - TAdv, Channels and Swings Vadimchi


Tadv - Adverse Tactics, presented by multi-points. Screenshots were posted here by one of the authors.

Vadimcha's channels and swings are models published online by a user called Vadimcha. I won't give any links, google the nickname, you'll find them. I searched for them myself three years ago.

Screenshots with models I posted just above.

My interest is in formalization of these patterns for automation.

This is what I would like to formalize. The screenshot shows the sequence of two black pulses. The task is to TI methods to DECLARE the length of the third red pulse and the moment of arrival at the end point.

 
Avals:

For there to be a profit, there has to be a directional move. I.e. this pattern signals a point at which quite a few people will start buying or selling on the market. What can drive them to do that regularly, in any market and forever)?

There is always directional movement. The sizes are different.
 
VNG:


TAdv - Adverse Tactics, presented by multipoints. Screens were posted here by one of the authors.

Vadimcha's channels and swings are models published online by a user called Vadimcha. I won't give any links, google the nickname, you'll find them. I searched for them myself three years ago.

Screenshots with models I posted just above.

My interest is in formalization of these patterns for automation.

This is what I would like to formalize. The screenshot shows the sequence of two black pulses. The task is to calculate the length of the third red pulse and the moment of arrival to the endpoint using THI methods.


you recently claimed that Vadimych's TA and patterns are the only ones that work 100% in the market. How did you verify this?