From theory to practice - page 613
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
Here I read the article Hearst index distribution of non-stationary labelled time series: http://www.keldysh.ru/papers/2013/prep2013_11.pdf
Excerpt: "Hearst's figure is therefore often used in financial market analysis to estimate the length of trends or to estimate the sample length over which moving averages should be calculated."
Yes, well, the Hearst index also depends on sample length....
Excerpt: "Hearst's figure is therefore often used in financial market analysis to estimate the duration of trends or to estimate the sample length over which moving averages should be calculated."
alas, this is the internet and there is a lot of writing especially about analysis of financial series, such "works" periodically appear on Habra, the structure of articles is simple: we take the first "that came to mind" mathematical method and apply it to financial series - then the conclusion hurray works! well and or trivial works, but not always..... like in this thread ))))
here wrote his visionhttps://www.mql5.com/ru/forum/86386/page1021#comment_8096557
i don't know why, but i have a feeling that in runet the mathematical apparatus is rather outdated. i googled SSA-method and partly read the information on english sites and then again in runet. but i got the impression that in runet they are translations of the english originals..... maybe in runet the matrix apparatus is something that was developed 20-30 years ago? and there are more modern methods for working with large amounts of data? and as a derivative - work with time series?
Voland's ball took place in a five-room flat.When Koroviev took Margarita to the ball, she said :
"... what amazes me most is where it all fits." She waved her hand, emphasising the immensity of the room. She flashed a hand to indicate the vastness of the room and Korovyov grinned sweetly, a shadow flickering in the folds of his nose. "The easiest of all," he replied, "for one who knows the fifth dimension well enough to stretch it to the desirable limits. I'll tell you more, Madam, to who knows what limits!"
М. Bulgakov "The Master and Margarita"
Dumbed down, in short.
It seems to me that the reading interval of the quotes (and maybe something else.) must change dynamically with the time of day.
You had a dynamic window and "ears".
Nah, Eugene, I think it's destiny to bring that bimodal distribution to the grail. Or to market the way to get it. For there's mystery and mysticism there, and hence the coveted treasure.
For there is mystery and mysticism, and therefore coveted treasure.
Alexander! What's the shift in the ACF calculation, what do we compare it to?
Well, in my WisSim it counts like this:
i.e. it is the sum of the products of the increments in the sliding window.
If the correlation is <=0 there should be an unrestrained pullback to the mean, if it is >0 there should be a continuation of the directional movement.
But, it doesn't work well for me.
The thing is that on tick data some pairs almost always have correlation coefficient >0, and I have to go to Erlang flows of higher and higher orders... It's torturous and slow moving...
Now I'm not surprised that commoners like Bas have been flipping coins for decades to concoct a peasant's Grail... It's a long, painful process...
What we need is an immediate Grail and unrestrained cash.
Perhaps you are ripe for perception:
( #1, #2, #3, #4 )
https://www.mql5.com/ru/forum/70676