This indicator is designed to demonstrate the potential of cluster filters and is a supplement to the article “Creating digital filters nonretarded time.” It contains a set of tests, through which the individual nuances of behavior of this type of filter can be explored.
All tests are made on the basis of the Momentum indicator.
Method – method of calculating the Momentum:
- Relative Momentum – Relative Momentum, the calculation is done as in Uilyama Blau. Installed by default.
- Absolute Momentum – the absolute momentum, the calculation is carried out similar to the built-in momentum of the terminal.
- Price Rate of Change – calculated as the ROC indicator.
All the indicators line identical and differ only in the vertical axis scale. All the descriptions of tests will be made for the relative momentum.
max bars – the number of bars for the calculation. Default 1000.
Filter – operation built a clustered filter:
- Off – the filter is disabled.
- Test No.1 Advance – the filter setting for the first test, “Effect of advanced anti-aliasing”.
- Test No.2 Smoothing – the filter setting for the test number two “Smoothing without delay”
- Test No.3 Impulse – setting for the study of the impulse response of the momentum.
During the second and third test, you can change the sensitivity of the anti-aliasing algorithm pressing the Up and Down with the active display window.
Test 1 Effect of advanced anti-aliasing
For the first demonstration of the need to set the “Filter” in the “Test No.1 Advance” mode.
cluster filter settings to allow the first test often enough to see the effect, which can be called a leading. When operating in this mode, you will not see all of the original anti-aliasing line momentum. Filter seeks out areas where there is a chance to get the effect of advancing. Not surprisingly, it turns out not always, but often enough to be able to notice it.
At first screenshot shows one illustrative plots of the filter. It is worth noting that this is only the appearance of advanced filtering. This impression arises solely because of the presence of noise movement itself unsmoothed line momentum, not because the filter ahead of time.
Test 2. Smoothing without delay
In the first test, it was shown that when the original line momentum is wrong (noise) against the trend, it has the effect of advancing the baseline smooth line. It is clear that the momentum of the line can be wrong, and in the opposite direction, that is suddenly slip away in the direction of the trend, and even obscenely long stay there, although the trend has reversed. It would be logical if the filter will begin to slow down such movements.
You can see how the algorithm works in the filtration’s leading trend of errors. For this startup indicator must set the “Filter” in the “Test No.2 Smoothing”. cluster filter operation during this test is divided into two parts.
The short name of the indicator “GMomentum (Option 1, Option 2)” is displayed in the graph pane, in brackets there are two options. If the second parameter is set to -1, then an attempt is made to correct (flatten) the error, forward-trend. If the second parameter is equal to or greater than zero, then connect setting for the advanced anti-aliasing. Use the keyboard Up and Down buttons to change the sensitivity of the filter.
The second screenshot shows that, despite the sharp price movements and the original line of momentum, the momentum line smoothing takes place without delay. With sufficient smoothing obtain acceptable entry point.
By the way, to the flat areas are the most trouble indicators. Here, even when these options, some of them easily degenerate almost in a straight line. Theoretically, using this technique, it is possible to improve any of the existing indicators, without creating an additional delay.
Test 3 Study of impulse response
Research momentum with built-in filter characteristics can be very amusing. For example, the impulse response of a good demonstration of how and where the peaks on the momentum line using the settings for the second test disappear. For the third test, you must set the “Filter” in the “Test No.3 Impulse”. During the test, a single pulse is applied to each of 1024 bar. After starting the indicator locate the point on the graph. It should look as shown on the third screenshot.
At start indicator is disabled. Therefore, you will see two peaks in the blue and red lines. One peak is a single pulse and the time it is, the other directed in the opposite direction through a specified number of periods. That’s what the impulse response of “naked” momentum. Further, using the keyboard keys Up and Down The, gradually increase or decrease the sensitivity of the filter.
It can be seen that the filter gradually under the root “eats” the second peak and does not touch the first. Filter completely corrects what done momentum and exactly reconstruct the original picture: a single pulse in the open field. There is no delay. No distortion of the unit pulse amplitude and shape. What is it? Work perfect filter?
For more information, see “Creating digital filters nonretarded time.”
SSA Stochastic Limited Edition version ” stochastic oscillator “Algorithms using singular spectrum analysis SSA It is an effective treatment method…
Trend Sight Trend Sight indicator shows the percentage change in the moving average of the current bar to the next selected the bar and allows you to…
Dinamic Trend Matrix MT5 New dynamic trend indicator matrix analyzes the current price development, the strength of the movement, its speed and…
- Volatility Hunter
- ADX Plus
- Divergence DeMarker or any custom