Bill Williams and his strategies... - page 11

 
artikul:
By dynamic programming, for example )))

Bellman optimality principle?

;)

 
Demi:


You do not confuse the cyclicality of economic processes and Elliott Waves. Cycles in economics have been around since the 19th century. The longest cycles were recorded and highlighted by Kondratieff in the early 20th century and this economic crisis fits his theory perfectly. Moreover, in the chart he drew in the 20's the bottom of the economic global crisis falls just in 2015 - 2018.

But Eliot waves have nothing to do with it. They don't make economic sense, and neither do the waves themselves.

And the point of your posts?

Expand your thought...

;)

If Elliot has no waves - then what is he preaching?

Or are you talking about the unknown Eliot - who has nothing?

We are not interested in it.

 
Sorento:

What's the point of your posts?

Expand your thought...

;)

If Elliot has no waves - then what is he preaching?

Or are you talking about Elliot, unknown to us - who has nothing?

We have no interest in him.


Alas for me! Ralph Elliot preaches nothing! He's no longer with us.... Dead....

And I was just drawing your attention to the difference between cyclical economic processes and Elliot waves

 

Any TA theory works the more effectively the fewer traders use it to trade. And the less effective the more amounts of money are traded in the market using this theory.

That's why the number of Fibo levels is constantly increasing, and old ones stop working. That's why old ones start working again after fewer and fewer traders use them for trading.

Crowd psychology and Oedipus effect, can't be helped.

 
Demi:


alas for me! Nothing preached by Ralph Elliot! He is no longer with us.... Dead....

And I just drew your attention to the difference between cyclical economic processes and Elliot waves.

I didn't see it. cyclicality is not Fourier in its pure form, but an alternation. and an unformalized harmonic...

Where's the difference?

;)

 

Сами по себе волнушки мало что значат без многочисленных дополнительных инструментов, только благодаря которым волны в руках истинного артиста начинают работать. Один из таких инструментов - это Фибы. На "заведомо случайных рядах" волны есть, но Фиб - нет (точнее, есть, но не больше, чем допускает сама случайность:)). Именно на финансовых рядах Фиб оказывается значительно больше, чем это допустимо, если бы ряды были случайными.

As far as I know, the only statistically verified level is 100%. By the way, the next thread here is just about that.

There are waves on the "notoriously random series", but there are no Fibs (or rather, there are, but no more than randomness itself allows for:))

Can you imagine a statistical method that proves the probability of waves at Fibo levels on real price series is higher than other levels? Just don't reduce everything to the cubature of the circle or cluster multivariate analysis of swings of different scales in 11-dimensional space again. The result of such calculations is well known: yellow walls and pictures like this:

 
Sorento:

I didn't see it. cyclicality is not pure Fourier, but alternation. and unformalised harmonic...

Where's the difference?

;)


The cyclicality of economic processes has a clear economic rationale. There is such a discipline - Fundamentals of Economic Theory.

Eliot waves have no economic justification whatsoever.

There is no pure Fourier in economics. Fourier in pure form happens only in a textbook on theoretical theory and mathematical statistics. Economic processes are the collective behaviour of the crowd. You can't pull "pure" statistical processes over it.

 
Demi:


The cyclicality of economic processes has a clear economic foundation. There is a discipline called Foundations of Economic Theory.

Eliot waves have no economic justification whatsoever.

There is no pure Fourier in economics. Fourier in its pure form can only be found in a textbook on theoretical theory and mathematical statistics. Economic processes are the collective behaviour of the crowd. You can't pull "pure" statistical processes over it.

Thank you, Guru!
 
Sorento:

the Bellman optimality principle?

;)


Let's try to forget for a while everything we have read, watched and listened to as a result of being in the financial markets. And think about what we have when we open the terminal. We have information about the price P, volume V and market volatility (tension) F = f(P) / f(V). The time goes by - the price, volume and the market calm change. The speed of these changes is also changing. As we all know - everything comes to an end sometime. Price and volume will not reach the current +/- infinity, nor will the acceleration of rises and falls. So, one day the system will reach the next unstable equilibrium and we will be at the top or at the bottom of the price wave. The output will be, as usual, the number P, which does not say anything about the fact that the input information structure has changed into the similar to itself antistructure. So we do not need an analysis of faceless numbers P, V and F, but a search of optimal structure of self-organized information at a given time interval. And the question arises at once, does it have a measure at all? Well, they will say to me - but what about all sorts of indices and graphical constructions? So? And they all have different measurements and scale. RSI is beloved by everybody for its levels 20 and 80, Elliott Waves must have 5 or 3 impulse waves, the length of the shadow of a bull hammer (just listen to it) must be significantly longer than the body, the waves with different periods must intersect somewhere, etc.

So, gentlemen, what shall we measure with? Using trend-following sticks, fibs or parabolic? )))

 
C-4: Can you present a statistical method to prove that the probability of waves at Fibo levels on real price series is higher than other levels?

No, I can't. I have written about this before. The problem is that the algorithm for checking this statement depends on what we mean by a meaningful Fib level (a level that price will not pass). With the primitive approach (Fibs are only counted from a single swing) the check really doesn't validate the significance of Fib levels. The author of the book "Secrets of Elite Traders" has it (I forget his full name; he is the author of a famous wave program).

But the check can be made much more difficult by making it more realistic. At every given moment one must consider at least dozens of swings of different scales, look at the Fib grid that has been obtained, and make a clusterization of levels. The task is very difficult. I haven't solved it.

Demi: any TA theory works the more effectively the less traders use it for trading.

Unobvious, by the way. The strength of a level is determined not by how many people play it, but by the power of the financial power behind that level. Of course, this info is not for trivia like us.