Hello all

After reading this excellent post by David, “20 Week Cycle Dominance – 2 May 2014″ , I want to share with you some ideas running in my head for some time about the “dominant cycle” and about what I mean when I believe that a particular cycle oscillation is almost disappearing.

As David mentions in the post, “…The subject of cycle dominance is a fairly big one…”, I think also it is like that.

Please let me explain some ideas about the concept of dominant cycle.

When trying to understand it, I can imaging Hurst trying to explain the cycle behavior in a static graph printed in a paper, and it is reasonable to think that the starting point for cyclic analysis is to visually establish what Hurst called the dominant cycle, and it is like David says: “…a dominant cycle is the cycle that is most visually apparent on the chart, the cycle that creates the most obvious cycle shapes…”.

It is specially true when considering a static chart but when you do zoom out in “modern systems” the dominant cycle could be longer and longer each time, reaching such a level for considering it as a too longer in relation to the time frame we use to trade. Please correct me if I’m wrong, but I believe that Hurst said something similar to this: “the dominant cycle is the **longer** cycle that is most visually apparent on the chart, the cycle that creates the most obvious cycle shapes **in which we can identify at least 3 full oscillations**”

If it is like that, I think that the concept of dominant cycle depends only in the time frame and number of bars showed in a graph and it can fall far of interest for our trading porpoises .

Being at this point I would like to talk about the other obvious meaning of dominant, and I would say that it is the one which is better understood by everyone, **I would say that we must better talk about the dominancy (the condition of being dominant) of a cycle, in other words, we must better talk about the popularity of a cycle in the neighborhood of our trading cycle.**

And I think that this dominancy or popularity of a cycle is in direct correlation with volatility, and I think it makes sense.

Let’s see an example of what I mean. Let’s say that our trading cycle is the 20 day cycle, the SUT is up to two cycles, the 80 day cycle, which also marks the 8 trading opportunities, at the same time we have the necessary of at least 2 cycles bellow for a good analysis, the 10 and 5 days cycle, so I’m considering our ” cyclic neighborhood” from the 80 day cycle down to the 5 day cycle in this example.

I think this statement is obvious, **when volatility decrease the dominancy shifts from longer cycles to shorter cycles meanwhile when volatility increase the dominancy shifts from shorter cycles to longer cycles. When volatility decrease the oscillation of the longer cycle tend to almost “disappear” favoring better and more recurrent oscillations in shorter cycles and vice verse.**

I also think, as we all may agree, that when we have decreased the volatility too much, then, a extreme movement is ready for starting and that a the longer cycle is prepared to take control.

The big question here is how we can decipher if we are entering into a decreasing volatility scenario or vice verse. In this sense I believe that the mathematical approach has something much to say about this.

Best

José

NOTICE: I post here my answer to Jonathan’s comments bellow in order to include a picture, it is not allowed insert one when answering the post!

Hello Jonathan

First of all I would like to thank you for sharing your thoughts and for contributing to the discussion.

What I’m suggesting is quite complex and difficult to show, but I would try to make it clearer.

I labeled the post as “Beyond Hurst” because my intuition is telling me I’m in a region bit out about what Hurst said. In any case, I’m fully agree with the dominant cycle concept; if we take the 20 day cycle as the fix trading cycle of our strategy we my have the dominance of longer cycles distorting FLD interactions.

I’m introducing a new prism about having a “fix trading cycle”, instead of that, we may have a “dynamic trading cycle” depending on volatility.

This nebula came to my head when I started to use the intraday version of Sentient Trader 3 years ago, when using it, I liked to trade sometimes the 1 day cycle but others I preferred the 2 day cycle, or even longer cycles, depending of time and the market, crazy, isn’t it?. Well, what I didn’t realized is that I was desperately looking for the nicest FLD’s interactions; it is clear that nice interactions mean confidence in what’s next. For that time I couldn’t even realized what was happening or how to decipher when these “shifts” happen.

Just after that, David launched the 20 day FLD strategy giving to us a great amount of light and safety while trading, thank you David!. I must say that probably the 20 day cycle is the most stable trading cycle from any prism you may apply.

Once here, I was wondering what’s the meaning of a stable cycle or stable interaction, the answer came after doing some job with the mathematical approach, in my personal opinion the answer is “stable oscillation” which, in my opinion, is the most repetitive ans sinuous oscillation…. and this oscillation is what I think is dynamic depending of the volatility.

Please take a look to the chart bellow, the indicators bellow are representing the 80, 20 and 10 day oscillation. Meanwhile the oscillation of the 80 and 20 day cycle seems to be rapidly diminishing (disappearing), the 10 day oscillation is making nice and evident oscillation. I would say that if you trade that cycle you would be more efficient in your trading meanwhile if you stay in your fix trading cycle you may be waiting and waiting until the cycle restarts its popularity.

I think it is evident in the chart that this process is in direct correlation with volatility

Best

José

Hello José,

It is a nice but difficult discussion. I also would like to decipher when a cycle dominance rises as early as possible.

In my opinion, the dominance of longer cycles has to do with Pseudo-trend where amplitudes become higher. Hurst defined that this early indication would be the Periodogram (page 113 Section six workshop unit number one – Lesson 8), when short wave amplitudes becomes larger than the long waves. However, David mentioned that the Periodogram has proved to be not so reliable.

Another way to anticipate this would be the “bullishness measurement” that David presented in his blog. If you expect a powerful bullish M shape to come next, then it might an early indication of cycle dominance.

As to volatility decrease when shorter cycles are dominant, I am not quite sure about it because I also think that volatility may also be high during shorter cycle dominance and David mentioned it in the FLD trading Strategy (lesson 6).

By the way, could you please tell me the page of following statement (in which we can identify at least 3 full oscillations)???

Another thing, could you tell me how did you plot the amplitude of cycles using the MACD? I remember a post from you like that to me.

Thanks José,

Best

Jonathan

Hello Jonathan,

In my opinion the most effective way to observe the amplitude of a price wave is with a properly constructed bandpass filter. It will show you the exact amplitude of the wave and provide you with a clear indication as to whether it is expanding or decreasing. A highpass filter will show you the exact amplitude of the sum of all the shorter waves.

Although not as effective as a bandpass filter, Hurst’s Periodogram will provide one with a rough indication as to whether the amplitude of a wave is increasing or decreasing. You must keep in mind that the Periodogram is only showing the amplitude of an individual frequency, whereas a price wave is a band of modulated frequencies. I may be mistaken but it does not appear to me that the Periodogram in ST is the same as the Periodogram in the cycles course. That may be the cause of it being “unreliable.”

Even though I find David Hickson’s “bullishness measurement” discussion fascinating, I do not recall Hurst ever mentioning anything about cycle shape other than translation effects. I also use the “visually evident” test, among other things, to determine price wave dominance. Jose is correct in his statement that what is visually evident is a function of the sampling period used.

I find the 80 day wave to be the most consistent price wave in the stock indices. The cycle to cycle oscillations are fairly consistent and the amplitude remains relatively stable over long periods of time (it has shrunk over the most recent oscillation). The highs and lows of the 20 day price wave are closely correlated to David Hickson’s alphabet “FLD interactions” , although a bit more timely, providing textbook action signals for entering long and shorting the 80 day wave.

Look for my next post under “Stock Market Analysis” for some chart examples of the above.

Hope this helps.

William

Hello William and José,

I also thank you for sharing your opinions. I’ve been studing Hurst concepts for almost an year and that seems I still have a lot to catch up with you. It is really clear to me if you anticipate how the amplitude will behave, you’ll be one step ahead. It makes me more and more fascinated by cycles.

I wonder if either of you could teach me your techniques to measure amplitude. It seems that José uses a different method from William’s one. I’d be grateful to have at least one way.

Best Regards,

Jonathan Santos

I forgot to mention… I think the way to find out if we are entering into a period of high or low volatility is only by Classical Technical Analysis using either Bollinger Bands or Historical Volatility indicator. You could say Hurst Volatility Index, but you need two instruments to compare (a stock and index). If we are talking about the SP500, we don’t have another instrument to compare with because it is already an index.

That is my point of view.

Best

Jonathan

Jonathan, please find my answer to your comments in the main post. I have included it there in order to attach a picture. Many thanks