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
Steady states are flat tops during reversals or corrections. Trends are unstable states of transition from one flat to the next. Before a trend, a regular signal is amplified by the noise of the flat and manifests itself in abrupt, often momentary jumps from level to level.
How can we learn something practical from this?
P.S. For example, how can we extract only the random component (pure noise) from volatility to get a regular signal? Volatility is known to be an antipersistent process. Simply subtracting a constant from it will not work, as the signal is getting stronger during the trend. Detrend? And what, I wonder, is the amplification coefficient equal to?
Noise extraction, wavelets seem to be designed for it (I may be wrong).
Good luck.
It feels like it somehow resonates with the potential models, or rather my view of where and how to use them :).
And where can I read about potential models from your point of view, because Google is choking on "potential models", life as always revolves around the littles.
There's a great thread on the parallel forum :) https://www.mql5.com/ru/forum/50458 but it's pretty much muddied. As an example see http://forex.kbpauk.ru/download.php?Number=16275. As for my view, such models just should describe steady states thus opening a fundamental possibility to isolate an "external" signal.
And where can I read about potential models from your point of view, because Google is chock full of "potential models", life as always revolves around lollygagging.
There's a great thread on the parallel forum :) https://www.mql5.com/ru/forum/50458 but it's pretty much muddied. As an example see http://forex.kbpauk.ru/download.php?Number=16275. As for my view, such models just should describe steady states thus opening a fundamental possibility to isolate the "external" signal.
If I understand the point correctly, you need to look for characteristics of the System/Model (maybe based on a time series) where this "current" weak regular signal will resonate with the noise, i.e. it will be multiply amplified. Specially highlighted:
I can be more specific: in other words, if approached from a practical point of view, one should control the parameters of the noise and look for such values of its characteristics at which the probability of resonance is significantly increased.
It will hardly be possible to calculate the exact trajectory, but it may be possible to calculate the main characteristics of future directional motion (momentum, jump, swing - whatever). Accordingly it is necessary to set:
- Parameters of the current noise (I assume they vary)
- Current signal parameters
There are, of course, some difficulties with this. The signal-to-noise relationship is strongly influenced by each other. For the current signal two simple options are in order:
- is to isolate the signal with some kind of low-pass filter (the wavelet option is quite good for this model).
- Use of various regressions or their combinations
In general case we will need to make a prediction of the same system elements:
- Parameters of future noise
- Future signal parameters
Prediction of noise will probably make you smile, but I think it should be an important part of the system. Of course you don't need to predict the noise itself but you do need to draw some conclusions about the basic parameters of the future noise. The resonance itself seems to me to be of a very random nature, and whether it will or will not add up depends almost entirely on the noise.
PS 01: this is an interesting idea, so it will take a year or more, taking into account the necessary research and trying out different variants.
to MathematI use this concept in my model. It works very well.
If I understand the point correctly, you need to look for characteristics of the System/Model (maybe based on a time series) where this "current" weak regular signal will resonate with the noise, i.e. it will be multiply amplified. Specially highlighted:
I can be more specific: in other words, if approached from a practical point of view, one should control the noise parameters and look for such noise characteristics at which the probability of resonance is significantly increased.
Calculating an exact trajectory is unlikely, but it may be possible to calculate the main characteristics of future directional motion (momentum, jump, sweep - whatever). Accordingly it is necessary to set:
- Parameters of the current noise (I assume they vary)
- Current signal parameters
There are, of course, some difficulties with this. The signal-to-noise relationship is strongly influenced by each other. For the current signal two simple options are in order:
- is to isolate the signal with some kind of low-pass filter (the wavelet option is quite good for this model).
- Use of various regressions or their combinations
In general case we will need to make a prediction of the same system elements:
- Parameters of future noise
- Future signal parameters
Prediction of noise will probably make you smile, but I think it should be an important part of the system. Of course you don't need to predict the noise itself but you do need to draw some conclusions about the basic parameters of the future noise. The resonance itself seems to me to be of a very random nature, and whether it will or will not add up depends almost entirely on the noise.
PS 01: this is an interesting idea, so it will take a year or more, taking into account the necessary research and trying out different variants.
to MathematI use this concept in my model. It works very well.
If I understand the article correctly, you have to look for a permanent source of influence. But it may turn out that there isn't one. Or there are many of them, which is the same thing. So how do you go about it?
If I've understood the article correctly, you need to look for a permanent source of impact. But it may turn out that there isn't one. Or there are a lot of them, which is the same thing. So how do you go about it?
I have a strong feeling that I need to look for both, and all together it's very frustrating. I'll start my journey with noise, especially as I've wanted to deal with it for a long time.
If I understood the article correctly, you need to look for a permanent source of influence. But it may turn out that there isn't any. Or there are many of them, which is the same thing. So, what to do?
I have a strong feeling that I need to look for both, and all together is very frustrating. Optional, I'll start my journey with the noise, especially as I've wanted to deal with it for a long time.
Of course, you could split up the tasks. But then you have to look for an answer - Who benefits? But that sounds like a childish question. Although I could be wrong.