Hearst index - page 14
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
Everything has already been stolen before us, yay.
The author of that indicator didn't see the benefit either ))
I've had a blast today. The analog of Hurst's coefficient is possible to calculate quite locally!!!!!!!!!
This follows from Dubovikov's paper "Minimum coverage dimensionality and local analysis of fractal time series"
finiplfbresibfjnuszrnmyfiscduuxawumosojpyvlqebhurzusezlkwygomdpegmoywhnwojmoacxeniugtkoxydf.rar
I enjoyed the article.
Thanks for the useful information. I would like to experiment... >>Thank you, my NS finds all sorts of patterns in kotir by itself.
The author of that indicator didn't see any use either ))
I didn't see anything useful in terms of trading.
The variation index is calculated on the previous interval and characterizes the stability of the process in the past.
There is no additional information whether the series will save its state in the future.
Therefore its application, e.g. as a criterion for changing the tactics of the TC is questionable.
Additional doubts were raised by MathCad research for integrated CB.
The variation index for such series is always less than 1/2, which most likely indicates
persistence of the series rather than randomness. And this contradicts the original condition.
....
The variation index for such a series was always less than 1/2, which is more likely to indicate
of the persistence of the series (which keeps the trend), rather than randomness. This contradicts the original condition.
What condition? Can you give us your opinion about it in more detail? Thanks in advance.
What condition? Can you elaborate on your thoughts on this? And if you don't mind, attach a Matkad file. The one shown above. Thank you in advance.
That the input series is NE (integrated).
If you apply a non-integrated CB as the input series, the situation does not change.
The variation index becomes greater than (but not equal to) 1/2.
P.S. File attached to previous post.The fact that the input series is SV (integrated).
If a non-integrated SV is used as the input series, the situation does not change.
The variation index becomes greater than (but not equal to) 1/2
So that's great, the best news in so many years. It turns out that all conclusions based on this premise (that the market is identical to an integrated random variable) are wrong. This is the second proof, somewhere here on the forum I posted (made the same conclusions), but based on the ACF look of the quotes. Hence it not martingale ('What is martingale?' which excludes theoretical possibility of earning), and it is possible to consider that theoretically it is proved possibility of earning on the market. It's not bad, it's great, just a little more ... )
Everything has already been stolen before us, yay.
Am I alone in sensing a certain slyness in the phrase "a restatement of the well-known market adage that movements of
of stocks or currencies are quite similar, irrespective of the time scale and price. The observer cannot tell from the appearance of a chart whether
the data refers to weekly, daily or hourly movements".
These are completely different things, weekly, daily and hourly.
Consequently, it is not a martingale ('What is a martingale?' which rules out the theoretical possibility of making money), and one can consider that the possibility of making money in the market has been theoretically proven.
There is no doubt that the market is not a martingale! To do this it is enough to plot PCs for different TFs or correlation coefficients between adjacent samples in the series of the first difference for different TFs to see the clear difference from the integrated SV with zero MO, which is a true martingale. I do not hesitate once again to show the comparison of these values for EURGBP pair and integrated CB as a function of TF expressed in min:
There is a difference from the martingale (compare red - martingale and blue -EURGBP), moreover, this dependence keeps the antipersistence trend to the TF more than one day, and then turns into a true martingale - to make a profit on the TF above the days statistically is not possible! The question is different. We, as traders, of course, are not interested in what is the kotir - martingale-not martingale, it is important - one can make money on it or not! So, it turns out that exploitation of persistence-antipersistence only does not allow to exceed the level of transaction costs on all TFs and pairs. In this sense (taking into account overheads), the market is a true martingale. We need to look for other methods to detect hidden patterns in price series. The only option, it seems to me, is to use elements of AI in the analysis, which allows to identify non-linear relationships in BP and markedly increase the profitability of TC as a whole.
These are completely different things, weekly, daily and hourly.
I can assure you that you will not be able to determine the difference between the M1 and the weeks (e.g. for the EURUSD series) "by eye". But the use of PU, will show exactly the difference between the different TFs for this quote.
Huh, and if we set a neuronet on this thing... Human perception is a subtle thing...