• Home
  • Forums
  • Trades
  • News
  • Calendar
  • Market
  • Brokers
  • Login
  • Join
  • User/Email: Password:
  • 6:34pm
Menu
  • Forums
  • Trades
  • News
  • Calendar
  • Market
  • Brokers
  • Login
  • Join
  • 6:34pm
Sister Sites
  • Metals Mine
  • Energy EXCH
  • Crypto Craft

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Planetary cycles, astrological theory of cycles of the markets 502 replies

Forecast Forex using Numeric Cycles and Astrology Cycles 36 replies

hurst cycles indicator 0 replies

Planetary Cycles... Plotting Cycles into the Future 0 replies

Using Lunar Cycles to create a Swing Trading Edge(In Stocks) 10 replies

  • Platform Tech
  • /
  • Reply to Thread
  • Subscribe
  • 1
Attachments: Idea for Hurst inspired indicator: Create cycles to predict price
Exit Attachments
Tags: Idea for Hurst inspired indicator: Create cycles to predict price
Cancel

Idea for Hurst inspired indicator: Create cycles to predict price

  • Post #1
  • Quote
  • First Post: Edited 9:29am Sep 29, 2017 8:12am | Edited 9:29am
  •  PipSharer
  • | Joined Oct 2009 | Status: Member | 50 Posts
Intro/purpose of post
As it says, I've got an idea to create an indicator.
The only problem is, my coding skills aren't great, so I'm wondering who would like to join me exploring this idea, and indeed develop some indicators for it?

Background
I was thinking that with much technical analysis we're taking price to create a single cycle in the chart window.
Ofcourse it can work well, but it is also limited. So how about we go in the opposite direction instead?

In other words, how about we estimate what cycles are operating, feed that data to MT4. And then ask it to predict a price according to that data.

My idea isn't revolutionary. It comes from Hurst waves. Hurst did a 1700 page course prior to the modern computer era.
Basically it explained how the wiggly lines we see can actually be explained by constructive and destructive wave interference. As an example, the classic double top pattern is simply one longer frame wave with 2 smaller regular sine waves (ie regular "up and down" waves) operating inside it.

The concept -how it would work.

1)Basics
What I propose is that the indicator allows us to choose certain waves. Ofcourse a wave can be measured as peak to peak or trough to trough. It doesn't matter as long as we are discussing a whole wave, as long as its the full "360 degrees". Its vital to remember that these are regular "up and down" sine waves. The only wave that is NOT regular, is the composite wave ie the price wave ie actual price or predicted price.

2)The fundamental of this concept.
So we might choose a wave that has a length of around one day, another of around one month, another of around one week, another of around 2 hours. Now if we define the length AND amplitude of each of these waves, according to Hurst we will get a "wiggley price line" that looks like a "real life price patttern".

What we would see is the price window. Below that would be individual cycle windows ie with an indicator window for each cycle.
(Remember each cycle is a regular wave, as discussed above).
And finally, in the price window we'd see the composite wave line plotted against the price, and running some way into the future ie the predictive price.

3)Addons/advanced aspects.
Ofcourse sometimes an external factor works on the market to give a larger move. For instance US Non-Farm Payrolls.
So perhaps we would have a system of weighting of waves in place eg the afternoon of the NFP's could have its wave given a higher weighting in the computation of the composite wave.

Likewise we might see bigger moves on market opens and sometimes on market closes. Again, a factor to consider with increased weighting of certain waves.

Please note that due to the composite nature of waves in Hurst theory, ofcourse we often find smaller waves repeating to give larger waves eg 4 hourly waves give one big 4 hourly wave turn and so on. In other words, the fractal nature of the market being demonstated.

4)Putting it together.
I wouldn't intend this as a standalone system. Support and resistance, classic price action, news events could all be added. As could many other systems. BUT this approach may give us a different insight into the market than the one we are currently using.

I look forward to replies, especially from anyone keen to do some coding. Thanks.
  • Post #2
  • Quote
  • Sep 29, 2017 9:47am Sep 29, 2017 9:47am
  •  Sixer
  • Joined Nov 2008 | Status: Member | 1,757 Posts
PS,

cycles are not "fixed" as Hurst assumed. So, take into consideration that his "Nominal Model" will not work.

Sixer
Attached Image (click to enlarge)
Click to Enlarge

Name: Hurst-Nominal_Model.jpg
Size: 45 KB
 
1
  • Post #3
  • Quote
  • Sep 29, 2017 12:08pm Sep 29, 2017 12:08pm
  •  PipSharer
  • | Joined Oct 2009 | Status: Member | 50 Posts
Good post. It would be interesting to see any statistical analysis on this. Although even that is of limited help. Because ofcourse there are almost infinite considerations.

I've never treated Hurst's table above as a "hard and fast" rule. So yes it is, just a "nominal model" ie rather arbitrary.
One of his fundamentals is that there is a fluctuation. For instance in a longer cycle trend, the smaller cycle fluctuations flatten out, sometimes so they are even inperceptible. And as I mention the benefit in weighting cycles, certain cycles are more important than others at certain times eg the cycles on the 15min chart at quiet periods vs the same 15min chart at NFPs time.

Additionally, the price action is affected by traders perceptions. For instance, exhaustion spikes when the market is turning etc.

I don't want to overcomplicate but you could argue we even have different types of cycle. News based, time of day based, purely technical and others. All we can do I suppose is choose a few of the key cycles. In the same way we only choose a few factors with neural nets to avoid "overfitting".

Traders spend a lot of time looking at multiple timeframes and think "so how does this all add together for MY timeframe"?
So my view is that this concept could help with an answer.
 
 
  • Post #4
  • Quote
  • Edited 6:08pm Sep 30, 2017 5:53pm | Edited 6:08pm
  •  dagoods
  • Joined Nov 2007 | Status: Member | 3,032 Posts
Quoting PipSharer
Disliked
Intro/purpose of post As it says, I've got an idea to create an indicator. The only problem is, my coding skills aren't great, so I'm wondering who would like to join me exploring this idea, and indeed develop some indicators for it? Background I was thinking that with much technical analysis we're taking price to create a single cycle in the chart window. Ofcourse it can work well, but it is also limited. So how about we go in the opposite direction instead? In other words, how about we estimate what cycles are operating, feed that data to MT4....
Ignored

360 degrees...sounds good. so long as we can get a slice of the PIE.

you can go crazy with the stuff

https://www.ams.jhu.edu/dan-mathofmusic/sound-waves/

https://www.forexfactory.com/showthread.php?t=534833

uses sound cycles not hurt...but whatever.... similar idea
https://www.newyorker.com/magazine/2...e-secret-cycle
 
 
  • Post #5
  • Quote
  • Last Post: Oct 2, 2017 12:22am Oct 2, 2017 12:22am
  •  dagoods
  • Joined Nov 2007 | Status: Member | 3,032 Posts
https://brilliant.org/wiki/double-slit-experiment/

way beyond LOL
 
 
  • Platform Tech
  • /
  • Idea for Hurst inspired indicator: Create cycles to predict price
  • Reply to Thread
0 traders viewing now
Top of Page
  • Facebook
  • Twitter
About FF
  • Mission
  • Products
  • User Guide
  • Media Kit
  • Blog
  • Contact
FF Products
  • Forums
  • Trades
  • Calendar
  • News
  • Market
  • Brokers
  • Trade Explorer
FF Website
  • Homepage
  • Search
  • Members
  • Report a Bug
Follow FF
  • Facebook
  • Twitter

FF Sister Sites:

  • Metals Mine
  • Energy EXCH
  • Crypto Craft

Forex Factory® is a brand of Fair Economy, Inc.

Terms of Service / ©2023