Searching for market patterns - page 38
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
It's an interesting comparison. Meanwhile, behind every such pebble is someone else's losses and unfulfilled hopes...
1. without a tick (equal volume) history, spectral analysis and spectral synthesis are almost meaningless. Above H1 it works fine, but the lower it is the worse the results from parasitic modulations.
2. To do normal spectral synthesis you need to solve 3 problems.
2.1 How to get rid of the amplitude modulation of the low-frequency part of the spectrum?
2.2 How can I predict and recover high frequencies which don't make physical sense on the graph? I.e. with a period smaller than 4 bars.
2.3 How to extrapolate the zero harmonic? Full decomposition into all harmonics again? So there will be zero harmonic again. And again again to all harmonics? Recursion? Good thing it's not infinite yet.
By solving these 3 problems you can get a very accurate bar forecast for 100 bars. Accuracy within the size of a bar. Bars, of course, are equal in volume.
I'm afraid you don't understand. Yes and how could you understand if I haven't said anything yet.
In general I am against calling spectral analysis or synthesis what I do. Because I don't work directly with spectrums and I don't see the need for it yet. It's just digital filtering and nothing more.
Somehow none of the 3 problems have ever bothered me, but there are other problems, but they seem to be solved. We all seem to have a very different understanding of it.
I don't quite understand what a 100 bar forecast is with accuracy within the bar size. You mean the accuracy is the same on all those 100 bars? Then what happens afterwards, why can't we do 200? My accuracy is the less accurate the further into the future.
There are 2 things that bother me about your approach(although I don't understand it much yet):
1. Where do you get a reliable tick equivolume history? When I was puzzling over the synchronization of currency pairs in time, I studied the historical data of many brokerage companies. They all have a lot of gaps even on 5-minute bars, especially USDJPY has gaps on 15-minute bars, as if there was not a single trade or tick. And of course the data on small timeframes is very different, and the tick history is completely different, although on large timeframes there is virtually no difference.
2. Your level of training in programming scares me. Is it necessary to implement such ideas? If so, they will be worse than useless for most, myself included. My suggestion is to keep it simple, to extract as much useful information as possible from the data that is available to everyone.
Don't be an obstacle to price movement... ))))
I'm afraid you don't understand. And how could you understand if I haven't said anything yet.
In general, I am against calling spectral analysis or synthesis what I do. Because I don't work directly with spectrums and I don't see the need for it yet. It's just digital filtering and nothing more.
Somehow none of the 3 problems have ever bothered me, but there are other problems, but they seem to be solved. We all seem to have a very different understanding of it.
I don't quite understand what a 100 bar forecast is with accuracy within the bar size. You mean the accuracy is the same on all those 100 bars? Then what happens afterwards, why can't we do 200? My accuracy is the less accurate the further into the future.
There are 2 things that bother me about your approach(although I don't understand it much yet):
1. Where do you get a reliable tick equivolume history? When I was puzzling over the synchronization of currency pairs in time, I studied the historical data of many brokerage companies. They all have a lot of gaps even on 5-minute bars, especially USDJPY has gaps on 15-minute bars, as if there was not a single trade or tick. And of course the data on small timeframes is very different, and the tick history is completely different, although on large timeframes there is almost no difference.
2. Your level of training in programming scares me. Is it necessary to implement such ideas? If so, they will be worse than useless for most, myself included. I suggest keeping it simple, extracting the most useful information from the data that is available to everyone.
The distribution of accuracy has not been specified. Of course, the further out, the lower the accuracy.
I'm taking the teak from Dukascopy.
I don't understand about the programming level etc.
it is hard to put into words ... the flow in the Euro, how do you see it happening ... I think virtually all Euro pairs will reflect this movement ... I think the beginning and the end of the movement may differ from one pair to another ... but in the end it will be a general move (when all the pairs are saturated) and each pair will finish this move at their own time (it may happen differently - we should try to catch the beginning of a general move, and the first pair that finishes it (and the others will continue for some time) may say that the move has begun to end .... that's the abstract way I see it
of course the infusion dose (or signal if you like) is different for each pair in the cluster - a kind of weighting probably... and because of the threat that many small arbitrage situations may arise, the infusion rates should be more or less even. I could be wrong about this.
what do you think about this..............?
I do not think it is the fault of the one who resists (perhaps it is not resistance at all) (perhaps it is simply a misunderstanding, or sometimes a fear of being misled unintentionally or intentionally (which is more common in our worldly dimension, which is far from being pure)).
Then... for starters... I suggest everyone accept and understand that the phrases ' it doesn't work - already discussed' should be shot on the spot, then doused with acid or lye, and then burned with a red-hot metal by sticking its pointed parts into the flesh.
Yusuf I would recommend you not to read and waste your time on this forum, go deeper into individual research. You are absolutely right when you talk about chaos at ticks and minutes, only I will add more at hours, days, weeks, etc., here it is important to understand one thing, on what scale to consider it, (here we could send you to Mandelbrot), it is only at first glance, a random movement does not carry any useful information - here they call them noise - gentlemen, noise only in your heads, urgently get rid of it )))))
Yusuf understand one thing - chaos at a micro-level produces new orders at a macro-level, this is not a pseudoscientific phenomenon, this phenomenon was studied by a brilliant chemist, Mr. Prigogine (do not get carried away by it too much), the famous Benar vortices, this process is observed not only in physical, chemical and biological systems, but also in social ones. Once you understand the mechanics of price movements, you will discover a lot of regularities and stationary throughout the market!
Good luck! :-)