Comparison of two quotation charts with non-linear distortions on the X-axis - page 3

New comment
 
The rule works on historical data: трейдеры охотнее продают чем покупают, the corners of ZZ of lower nodes are statistically sharper than corners of upper nodes of ZZ

Only questioned the red statement. Not the green one. I think the logic from the results of the green fact leading to the red assertion is flawed.

ZZ itself can be constructed on the basis of various kinds of conditions. Classical ZZ is an unsubstantiated idea at all. Measurement of angle on the basis of a ZVR bar filter is strange again.

But there is a positive side: symmetry (in context of statements above) for tops and bottoms is not a bad criterion for evaluating the adequacy of LA. If the symmetry is preserved, it means that the logic is adequate. Otherwise it is not.

 
Integer:
Zigzag. A series of values on the tops. Correlation of these two rows.


This is the first thing that came to mind. It raises strong doubts about applicability due to extra vertices even on very similar graphs.

DTW looks the most promising so far, need to code it into mql4 and try it out.

 
wmlab:


This is the first thing that came to mind. I have strong doubts about its applicability due to extra nodes even on very similar charts.

So far DTW looks the most promising, I need to harden it into mql4 and try it.


Take a deeper zigzag. What is DTW? Is it just a scaling along the time axis? But it's uniform. If it's a satisfactory solution, then there shouldn't be any question, it's the first elementary solution.

 
hrenfx: But there is a positive side: a good criterion for evaluating the adequacy of a restraint is the symmetry (in the context of the statements above) for the top and bottom vertices. If the symmetry is preserved, then the logic is adequate. Otherwise it is not.
Hm, of course, I would like to see at least some symmetric indicator, but I doubt that there is any - the tops of this ZZ used https://www.mql5.com/ru/code/10041, ZZ_FF_v4.mq4
 
Integer:


Take a rougher zigzag. What is DTW? Is it just scaling along the time axis? But it's uniform. If it's a satisfactory solution, then there shouldn't be any question, it's the first elementary solution.


As far as I understand the algorithm, it is exactly a measure of the similarity of two series in the presence of non-linear distortions in time. It is used in comparing audio patterns.
 
Yes, DTW has not yet been implemented on mql, although it is an option. And maybe it makes sense to look through the equivolume and Range bars, i.e. to abstract from the time component. Just find the optimal ratio of volatilities and compare them by increments.
 
sayfuji:
Yes, DTW has not yet been implemented on mql, although it is an option. And maybe it makes sense to look through the equivolume and Range bars, i.e. to abstract from the time component. Just find the optimal ratio of volatilities and compare them by increments.

I wrote in the thread above that Renko and the company are not suitable because of the strong amplitude noise of the quotes.
 
IgorM:
Hm, of course I would like to see at least some symmetric indicator, but I doubt if there is any...
  1. Logarithm the original CVRs (Bid and Ask).
  2. ZZ is constructed with the condition that the knee size >= N and the number of knees is maximal. Top tops lie on Bid, bottom to Ask.

I did not check for symmetry criterion.

 
hrenfx:
  1. Logarithm the original CVRs (Bid and Ask).
  2. ZZ is constructed with the condition that the knee size >= N and the number of knees is maximal. Top tops lie on Bid, bottom to Ask.

Did not check for symmetry criterion.

OK, but it is not possible to estimate BP at tick level, the doubtful tick history in the open access network, as well as independent tick collecting is a labour-intensive activity, the result will depend on the quotation source. I don't even want to go on why I don't consider tick data now.

As far as I understood your opinion about asymmetry of TF - it's all because of incomplete data representation in the form of OHLC and absence of information in OHLC separately about Bid and Ask ? But why has the price been moving in one direction for a long time (trends)? I think the matter is not ticks and OHLC. The code for plotting ZZ by points, I posted at https://www.mql5.com/ru/forum/127093, ZZ_line.mq4 draws correctly on history - it converges to ZZ one by one, I did not deal with it for online work, try to analyze by yourself the slope ratio of ZZ lines, maybe I am mistaken

 

For the construction of a PP according to the condition described above, the tick history is not needed. It is enough to have the OHLC M1 Bid and Ask history. This history is available in many sources, including some brokers on the MetaTrader4 platform (by standard means).

But why does the price go in one direction for quite a long time (trends)?

That is the question of pricing. I will not be able to cover it.

The code for plotting a phase by points, I posted at https://www.mql5.com/ru/forum/127093, ZZ_line.mq4 draws correctly on the history - one point converges to a phase.

It correctly draws the classic ZZ - the indicator invented by someone who is not even able to show the best entries/exits on the history.

12345678910...12
New comment