Machine learning in trading: theory, models, practice and algo-trading - page 2324

 
mytarmailS:

The idea is that the market can be described by two sine waves + the trend is linear or also a sine wave.

The sinusoids must have the right frequencies relative to each other in order to get the classical Eliot model of 5-3-5... and so on in a circle.

Then all TA figures show themselves, head shoulders, double top etc... And there is no magic here

video - shifting phases in harmonicshttps://radikal.ru/video/oVpWCd1Q1pA

If we find a way to shift prices to the same form (with clear parameters), success is practically guaranteed.


Here is even steeper, already with a trend, clearly visible 3-5 structure...

If the trend is described correctly, then it is possible to do without the rest of the sinusoids at all))

Do you really think that two sinusoids, i.e. functions that do not change their parameters, can describe a MARKET?

Not a part of the graph chosen arbitrarily or not at all, but a MARKET? I.e. it is possible to describe the entire length of the market chart, well at least for 10 years, by two sinusoids?

 
Aleksey Mavrin:

1) Do you really think that two sinusoids, i.e. functions that do not change their parameters, can describe a MARKET?

2) Not some part of the graph chosen arbitrarily or not so much, but exactly the MARKET? I.e. it is possible to describe the whole length of the market chart, at least for 10 years, by two sinusoids?

1) no

2) no.

I was talking about something else, the rest is yours.

 
mytarmailS:

1) no

2) no.

That's not what I was talking about, the rest is yours.

I quoted the text you wrote in Russian. If you were talking about the other then you chose too categorical a form of assertion, which distorted any meaning. Maybe you can clarify.

 
Aleksey Mavrin:

I quoted the text you wrote in Russian. If you were talking about something else, then you chose a too categorical form of statement, which distorted any meaning. Maybe I should explain.

And you try to read more than one message and understand what you were talking about.

 
mytarmailS:

And you try except for one message, read the previous few to understand what it was about

Oh, thanks for the tip, that's what I was missing, I choose at random messages for a response without reading the rest ))

And seriously - you would be better to formulate clearly and scientifically what you want to say.

I assume that you essentially said - two sinusoids (well, and a trend) can approximate any part of a graph with some maybe even "decent" accuracy.

Is that correct, or was there some other underlying meaning?

If you know what you're talking about, there should be no problem explaining what you're saying!

 
Aleksey Mavrin:

Oh, thanks for the tip, that's exactly what I've been missing, picking at random posts to reply to without reading the rest ))

As funny as it is to you, that's exactly what you're doing...

Aleksey Mavrin:

If you know what you're talking about, there should be no problem explaining what you're saying!

Let me clarify...

I was saying that it would be nice to create some "converter" of market data (non-stationary) - into a model (stationary, simplified, visual, with preservation of the structure we need) to represent this model with sine waves


All scientists around the world do it, to understand a complicated process, build a model, study the model, predict the model but not the process itself, it's the world practice, everybody does it, except for the people with the lowest level of training who believe that AMO will do everything...

 
mytarmailS:

As funny as it is to you, that's exactly what you're doing...

Let me explain...

I meant that it would be good to create a "converter" of market data (not stationary) into a model (stationary, simplified, demonstrative, with preservation of the structure we need) and this model can be represented by sine waves


All scientists do it all over the world to understand a complicated process, build a model, study the model, predict the model but not the process itself, this is the world practice, everybody does it, except for the MO scientists with the lowest level of training who believe that AMO will do everything themselves...

The thought is correct, there is no need to prove anything to anyone, just implement what you already have an idea about. The market can be brought to a common form (ula), just on one instrument it will work once a year, and the other every hour.

 
You've written here about "noise" filters, filters are evil because no one knows exactly how much of a signal's probability is going to perform. Adding a filter will further reduce your chances.
 

MO is the same research tool, only much more powerful than classical statistical analysis.

Any model is analyzed and conclusions are drawn

It's funny to read this thread. Until you throw in some new material, it will stagnate to the level of orangutans. This is all because of ignorance of the econometric basis, and ignorance of anything at all :)

 
Why downgrade your worldview with outdated models while the rest of the world develops MO