Non real-time envelopes


#1

I would define these as Hurst’s envelopes, removed from the distorting effects of time.

Do you use them? Are they as useful as Hurst claims? If so, do you produce them by hand?

My first thought on reading this chapter is that he is speaking of ensuring that the succeeding higher lows of a bullish trend must be higher than the previous high, otherwise, we suspect price is due to ‘curve over’. This notion is the basis for many successful trend following systems.

Beyond this though, does the envelope have utility? Hurst claims that we can use it to extrapolate price targets, but is it more useful than using real-time envelopes?


#2

One needs to be quite skilled at curve-fitting to produce them. They are not produced by hand. The main difficulty is the fact that the highs today are not as well defined as the example in the book. Another very technical issue is the fact that the example curve in the book does not have second derivative continuity throughout making it difficult to determine his mode of fitting. I would recommend trying to reproduce something simple (p. 76 see chart below) and then the non-real time example in the book before trying on current data.

I use envelopes for price projections but never for action signals because they are moving targets at now time.


#3

I use envelopes as a starting point but definitely not for trading signals. One of the trading systems I use includes volatility-based envelopes (not constant width) that use centered WMAs and regressions to extrapolate through the current bar. I usually use two envelopes with a harmonic relationship of 4 to 1 based on visually evident lows. It sounds a lot fancier than it is. I find the visual context is sometimes helpful, but other times a little misleading. When a big jolt occurs (first week in Feb), daily envelopes of this type will be greatly distorted. Given the recent increase in volatility, I have been using envelopes based on 30m or 60m bars because the increased number of data points act as a volatility shock absorber. I also find this approach greatly reduces the end point effect (repainting). I will post a chart later.

In theory, envelopes are the perfect trading tool because they identify all the market inefficiencies traders attempt to exploit (undervaluation, up momentum, overvaluation, down momentum, and back to undervaluation) all in a multi-time frame context. In practice, it ain’t that easy, mainly because they don’t help you with action signals on the right edge of the chart. They look great in the rear view mirror though!

Look at some of the triad charts Alain posted a while back showing the intersection of multiple DMAs (maximum point of momentum) then imagine 2 standard deviation envelopes around the price action. If you strip everything down to the very basics, that’s all there really is to it.

Curt


#4

My envelopes aren’t as pretty as William’s but they work in a pinch as an eyeball confirmation tool. A cross of the shorter and longer envelope midpoints usually approximates where price crosses the FLD (envelopes, like the triad, pick up shorter swings too). If your phasing is on the money, the FLD is by far the most accurate tool IMHO. Look at this example using the 20D nominal FLD and the category interactions to date also make sense (3/23 as the midpoint of the 80D nominal cycle).

I’m guessing experienced Hurst analysts often skip the envelope step altogether. Precise phasing and an FLD based on your trading timeframe is the key to low-risk trade entries.

Hope this helps.

Curt


#5

It might seem silly but I’ve been using FLD cat interactions without paying mind to the overall cycle length or the approximate cycle degrees where each leg is expected to start/end. Now that I can assign an approximate length to each step it should make things clearer.

Hurst puts so much emphasis on envelopes in PM that I’m surprised they’re not a bigger part of ST.


#6

It does seem silly because not knowing the basics is a great way to lose money consistently. Then you’ll get frustrated and quit. That would be too bad because this stuff works.

Take the ST FLD trading course if you haven’t already. There is a precise relationship between the cycle you are trading, the FLD, and those category interactions.


#7

This is David Hickson’s course? Have you taken it? Do you recommend it?


#8

If you are a ST user you should have access to it. They reorganized their educational videos/webinars last fall and it is part of that. Yes, I would recommend it. There is valuable information you can pick up regarding underlying trend from the strength or weakness of the category interactions. I should go back and look at it again. If you aren’t a ST user, there is some basic stuff on the ST website that talks about the trading cycle, FLD, and category interactions. This and the FLD webinars on YouTube would be a good start.