Wave analysis - page 23

 
Well, at least some example, some way to confirm it.
 
HideYourRichess писал(а) >>
Well at least some example, methods of confirmation.

1.Take a weekly chart, decompose it into a Fourier series and see that the price is the sum of several statistically significant sinusoids.

2. Take economic statistics for "any" indicator and clearly see cycles. Then, if the fundamentals are chosen intelligently, a very good multivariate regression of price on the indicators can be made, i.e. price = sum of cycles (with coefficients).

 

In confirmation, the wave pattern of the weekly charts is very clear.

 

I warned you! )))

A sea of gibberish and... that's all. To a concrete offer to write a code, run it in the tester and present a report to an amazed and envious public, the Wavehunters respond zombie-like: Read the book.

Sometimes they cannot hold back and in a religious ecstasy start to quote it, retell it in their own words and show pictures. Then pride takes over again and again: Read the book.

As Gleb Zheglov would say if he were an MTS trader: The code is the only and unique proof.

I will tell you what happens next: It will be the same as the previous 22 pages.

There will not be a code. And if there is a code, they will try not to notice it, because the code written exactly according to the reasoning of the wave runners will kill their dream at the root. And if they will notice it, they will explain the result incompatible with their reasoning by its - the code - inconsistency. And if you ask them what is imperfection, the answer will be - well, guess what? - Right: Read a book.

That's it. The circle closes and the story begins all over again.

Happy wave groundhog day to you, dear comrades! Yay!
 
sak120 >> :

1.Take a weekly chart, decompose it into a Fourier series and see that the price is the sum of several statistically significant sinusoids.

Very solid looking highlighted phrase. How do you estimate the statistical significance of one particular sinusoid? Because we've been talking about Fourier many times and it's useless...

 
Mathemat писал(а) >>

The highlighted phrase looks very solid. How do you assess the statistical significance of one particular sine wave? We've been talking about Fourier's many times, but it's useless...

Write a regression of prices from sinusoids. Regression methods.

 
sak120 >> :

Write a regression of price from a sinusoid. Regression methods.

That's not an answer, it's an excuse like "back off, nasty: even a fool knows how to do it, and you're picking on me". Let's put a smiley face just in case, so no hard feelings. :)

OK, let's come at it from the other side, if you're too lazy to write an algorithm here. Name one statistically significant sine wave on the weeks. And show me by what criteria you were able to figure out that it was statistically significant. That is, how do you understand this clever phrase.

 
sak120 писал(а) >>

Write a regression of price from a sinusoid. Regression methods.

What is good about Forex, it is that you can blunder some clever phrase, mark on the history all waves and you can think that you almost automatically become an expert on Forex and get credit and respect........)))))

 
sak120 >> :

1.Take a weekly chart, decompose it into a Fourier series and see that the price is the sum of several statistically significant sinusoids.

2. Take economic statistics for "any" indicator and clearly see cycles. Then, if the fundamentals are chosen intelligently, a very good multivariate regression of price on the indicators can be made, i.e. price = sum of cycles (with coefficients).

1. In doing so, it will turn out that these very Fourier... - It has already been so much discussed that there is no stationarity there and that Fourier is generally not applicable, and instead of a normal expansion we get scientific nonsense - I don't want to remember.

2. Cycles are periodic, with phase and amplitude. And as I understand it, with a constant phase and a constant amplitude. I'd love to see such a thing on the market - it's a grail. No, a grail! That's right.

 
sak120 >> :

In confirmation, the wave pattern of the weekly charts is very clear.

And where are the cycles here? and where is the Fourier?