Ehlers Autocorrelation Periodogram tries to identify the dominant cycle.
Can I use that output as the lookback period for other indicators?
Seems to me, having a given indicator in sync with the current cycle will be
quite useful.

I've looked around and I can't find anyone that's even tried this, which seems odd.
My preference is to use the Periodogram for Ehlers Sine Wave. I've seen the Sine
pick turning points with great precision and other times it's off by a LOT.
Logically, if the Sine Wave is always in sync with the dominant market cycle then
I like my chances of getting good info from it.

Ehlers original version plots a line (see "NT8_Periodogram_Line) as in this
TradingView example.

The NT8 version doesn't plot an exact line value and at times it's confusing.
See "NT8_Periodogram_Confusion"

I have tried plotting the Periodogram and several Sine Waves of various lookbacks.
When the Periodogram is near one of the Sine Waves, it does give better signals.
This method is a lot of work though and it does tend to take me away from other
aspects of my trading.

So, has anyone used the Periodogram to vary the lookback input for indicators?
Is anyone aware of why this approach would fail?
I thank you in advance for any help, JM

johnnymustard 

Ehlers Autocorrelation Periodogram tries to identify the dominant cycle. Can I use that output as the lookback period for other indicators? Seems to me, having a given indicator in sync with the current cycle will be quite useful. I've looked around and I can't find anyone that's even tried this, which seems odd. My preference is to use the Periodogram for Ehlers Sine Wave. I've seen the Sine pick turning points with great precision and other times it's off by a LOT. Logically, if the Sine Wave is always in sync with the dominant market cycle then I like my chances of getting good info from it. Ehlers original version plots a line (see "NT8_Periodogram_Line) as in this TradingView example. The NT8 version doesn't plot an exact line value and at times it's confusing. See "NT8_Periodogram_Confusion" I have tried plotting the Periodogram and several Sine Waves of various lookbacks. When the Periodogram is near one of the Sine Waves, it does give better signals. This method is a lot of work though and it does tend to take me away from other aspects of my trading. So, has anyone used the Periodogram to vary the lookback input for indicators? Is anyone aware of why this approach would fail? I thank you in advance for any help, JM

Ehlers published a book - 'Cycle Analytics for Traders' copyright 2013. In chapter 11 on "Adaptive Filters" he discusses an "Adaptive RSI"
indicator in which he uses the autocorrelation periodogram to determine the Dominant Cycle which he then uses to calculate "MyRSI".
"MyRSI" is an RSI that is tuned to half the dominant cycle. A similar approach is used to calculate an "Adaptive Stochastic Indicator" and an
"Adaptive CCI" and an "Adaptive Band-Pass".

Since the book was published Ehlers has published many articles and may have refined his techniques. You may be able to contact him for more information.

One thing to keep in mind is that as per Ehlers in the cited book "market cycles are evanescent - they come and go and change their periodicity over time". That could have implications
for your trading and use of this approach.

Also, I think that this approach implicitly models the market as having a "trend" with a cycle component superimposed. By limiting the analysis to a dominant cycle (which undoubtedly makes the problem more tractable) and filtering out frequencies
that are too high or too low one might be ignoring other cycles that could influence the ultimate result.

Good Luck!

TraderDoc2 

Ehlers published a book - 'Cycle Analytics for Traders' copyright 2013. In chapter 11 on "Adaptive Filters" he discusses an "Adaptive RSI" indicator in which he uses the autocorrelation periodogram to determine the Dominant Cycle which he then uses to calculate "MyRSI". "MyRSI" is an RSI that is tuned to half the dominant cycle. A similar approach is used to calculate an "Adaptive Stochastic Indicator" and an "Adaptive CCI" and an "Adaptive Band-Pass". Since the book was published Ehlers has published many articles and may have refined his techniques. You may be able to contact him for more information. One thing to keep in mind is that as per Ehlers in the cited book "market cycles are evanescent - they come and go and change their periodicity over time". That could have implications for your trading and use of this approach. Also, I think that this approach implicitly models the market as having a "trend" with a cycle component superimposed. By limiting the analysis to a dominant cycle (which undoubtedly makes the problem more tractable) and filtering out frequencies that are too high or too low one might be ignoring other cycles that could influence the ultimate result. Good Luck!

I appreciate your take on the matter, thank you. JM