Market model: constant throughput - page 27
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
I can add another interesting task of the same nature:
- bring the data to the same scale
- add up
- task-divide and get the original signals
I think what is meant is this: a starting point is selected. From this point the compression of the information starts.
I.e. 1, 2, 3, 4, 5 etc. bars are taken. The amount of information after the compression is shown as a chart.
Perhaps... there will be interesting curves in some points of the chart
I don't understand the method.
You choose some algorithm for obtaining information from the ones demonstrated. You fix a constant of information. Then, at some point, perform the following iterations
(1) You stand on the "current bar". From this "current bar" you successively take segments (increase the length of the window in the history) and for each segment count the amount of information according to the adopted algorithm. As soon as you find a segment with <constant> +/- <delta> - you write the length of this window for the current reading.
(2) You move to the next bar and repeat the search.
As a result, you get a series with lengths of BP bars, for which the condition - a fixed information constant (conventionally - "channel") is met. Collect statistics on these "channels" because while going blind collect everything that will be collected (RMS, ACF, etc.).
Next, you will need to investigate the resulting "channels". For example, choose a model(linear regression, correlation models, etc., at least TA) and look how the future BP is relative to the current bar on this model (for a start).
Suddenly something will turn up.
I think what is meant is this: a starting point is selected. From this point the compression of the information starts.
I.e. 1, 2, 3, 4, 5 etc. bars are taken. The amount of information after the compression is shown as a chart.
Perhaps... there will be interesting curves in some points of the chart
Can I have any example of this?
Put two frequencies together, 440 Hz and 500. Put it in Fourier.
Fourier gave two peaks.
Generated two separate components that give the original.
Understood, thank you. Implementing such a method would require much more computational resources than the method whose results have been presented.
If you set a small window - then the calculations are not big. An example is attached.
Easy :-D
Put two frequencies together, 440 Hz and 500. Put it in Fourier.
Fourier gave two peaks.
Generated two separate components that give the original.
I see, thank you. The preparatory work (compression) of this kind of method is very time consuming. I'll try it as soon as enough computing resources are available.
You need a starting point (from a vertical line) and a window - how much to count.
If you set a small window, the calculations are small.