Similarity: A Practical Approach | Page 10 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Similarity Trading - The ZZBB Method 215 replies

Rainbow similarity 256 replies

Help implementing zigzag similarity concept indicator 0 replies

Similarity Stuff 17 replies

Directional Similarity 0 replies

Attachments: Similarity: A Practical Approach
Exit Attachments

Similarity: A Practical Approach

  • #181
  • Jul 29, 2013 1:58am Jul 29, 2013 1:58am
Quoting Dewey McG
Thanks for starting this thread. Here is something I was working on before getting locked out. I have been working on 2 strategies that came from different ideas that were presented which I think show some promise: Strategy 1 Stoch goes outside band EMA goes outside same band ZZ Top or Bottom forms Enter trade SL = 10 pips TP is middle band Strategy 2 Stoch goes outside band EMA goes outside same band ZZ Top or Bottom forms Price crosses mid band enter trade SL is opposite band (top band if a short, lower band if a long) TP is next ZZ top or bottom...
Hi,

I was thinking of strategy 1 too. Looks full of promisses. Use the mean and extremum points.

Then i go back to the definition of the bollinger bands and i swap my interpretation : IF the price makes almost tops and bottoms outside the bollinger bands in accordance with the deviation parameter, THIS IS BECAUSE the price now is almost a gaussian function around its moving average mean, SO THAT it i use the bollinger bands, it seems we are trading the NOISE of the market.

So, the next question was : is it good to trade the "noise" or to remove it ?

When we have dissimilarity, this "noise" is completly distorded and we don't have alternating high and lows anymore. Fractals indicator shows mutiple tops or bottoms, ZigZag repaints its extremum, stock makes multiple '100s' or '0s'.
Enter Signature
  • #182
  • Jul 29, 2013 2:07am Jul 29, 2013 2:07am
Quoting Kiads
{quote} Yes, that really correct. i use 2 bb (high and low) to reduce that problem..
Interesting! Would you like to explain that or maybe show a chart how it looks and how you use that method?
  • #183
  • Jul 29, 2013 2:08am Jul 29, 2013 2:08am
Quoting Dewey McG
I wasn't expecting to take any trades in the Asian session, but when I pulled up my chart this was right here in front of me. Since it is a demo I decided to go for it, even if my entry was a tad late. I also clicked the close trade just before it hit the middle band--trigger finger but you get the idea. {image}
Exactly what i am looking at actually, but i think this is trading the noise and actually the market is swinging rather than trending. This is not "a practical approach to similarity", even this is great work.
Enter Signature
  • #184
  • Edited 3:21am Jul 29, 2013 2:18am | Edited 3:21am
Quoting Kiads
{quote} Yes, that really correct. i use 2 bb (high and low) to reduce that problem..
I found better to enter at the mean of the bollinger band and exit at the extremum :

- Not sensitive to runaway at extremums : VERY GOOD.
- you ask or bid at an "average" price so you have a good chance the price will go at least away from its average
- works during ranging AND trending markets : VERY GOOD

Then for the direction of the price, i choose the alternation principle (after a top, a bottom. And vice versa).

So the idea is good only 97% of time because when we have dissimilarities, this alternation process is not working there. 97% of the time ??? No, because the noise is distorded by news, market openings and closings, etc...

But for me this is trading the "noise" and don't know if its better to trade it or to remove it...
Attached Image (click to enlarge)
Click to Enlarge Name: trade.jpg Size: 300 KB
Enter Signature
  • #185
  • Jul 29, 2013 2:56am Jul 29, 2013 2:56am
There was/is a thread here based more or less on that principle.
  • #186
  • Jul 29, 2013 3:26am Jul 29, 2013 3:26am
Quoting Kiads
{quote} Hi Metcalfe, sorry for late reply here my template and also u need that 3 custom indicators for complete setting the rest are standard indicators.. pass for #Signal_Bars_v10a ff : free_at_ff hope it helps Kiads. {file} {file} {file} {file}

Thanks!
  • #187
  • Jul 29, 2013 3:45am Jul 29, 2013 3:45am
Good morning guys,

With the BBZZ "method", let's not try and analyse it to the point of optimisation. If you were to do that, you'd need the complete DS Framework which we don't.
At best it is a method - a flawed but useful method of applying an observation by Eurusdd.

The "similarity principle" being applied here is that PRICE = STOCH. McG's 2 strategies are a reflection of where we were on the other thread at the time we were debating the things we are discussing now : Repainting of ZZ. Essentially, the 2 strategies are

a) OUTSIDE TOWARDS INSIDE : trade it from the extreme going in
b) INSIDE TOWARDS OUT : trade it from the BB towards extreme.

----
Someone made the comment that the stoch needs to be at extreme ends of its measurement spectrum. Correct!
If its scale is say 100 to 70 and it values at 85 AND is outside the BB.... I would say it's a week signal. If it's at 70-75 and outside BB (ie a ZZBOT) it's more powerful. Reverse for the other side.
----
Just because a ZZTOP/BOT appears on a 5M.... doesn't mean it's a signal to trade. If it resolves to a 15M signal, yes - potential trade setup (now look for your own entry signal). At this point, I don't have a full blown method that is shareable because everyone is different! Where do you stop? Some of you may know that I am trying to figure out a Daily Direction method.... I can tell you this relies on ZZ, BB, Price/Stoch and 2 extra indicators that haven't been talked about as much. That's it. Small sample size (4 days... 3 correct so far), but the results have surprised me. What this leads me to believe is that the resolution should start from the highest to the lowest timeframe you are comfortable trading in.

Eurusdd never explained why he uses 4500 minutes or 300 bars of 15M. But the 300 has also been used in his version 2 DS Zone indicator.... which he said could be applied multi-timeframe. So he is basically saying 300 unites of time is enough. The 14 days of expiry - he turned into 35. I don't know why but just be aware of these.

So let's not try and "automate" the BBZZ method just yet.... let's continue to debate it. But I hope we don't dumb it down to the point that just because you can't write an EA for it - we mentally discard it or create the impression that the principle isn't "systemis-able"

Peace! Out!
A pip is only worth it if you know how much you risked to earn it
  • #188
  • Jul 29, 2013 3:46am Jul 29, 2013 3:46am
Good morning guys. Here I post my thoughts / views and approach to work with sim (similarity) and dissim (dissimilarity) zones. It may be a little different from the rest here. I do not use MT4. As tools I use only BB and stochs. The zones you see are hand drawn.

sim zone = Price outside BB (close) and Stoch outside BB. As my zone I mark the first retrace candle. The zones are the grey boxes!
dissim zone = Stoch outside BB but Price NOT outside or Price outside and Stoch not. As my zone I also mark the first retrace candle. The zones are the blue boxes!

Clearing of zones: For Upper zone (Price and Stoch above BB) we need a close above and a close under to clear it! / On the downside we need close under and a close over the zone!

The zones with a frame aren't cleared till now.

You can see how the zones act as S/R too when price breaks the bb on the downside.

Hope this helps to have another view to sim/dissim. More later.

Attached Image (click to enlarge)
Click to Enlarge Name: 13.JPG Size: 235 KB
  • #189
  • Jul 29, 2013 3:52am Jul 29, 2013 3:52am
Quoting xixi
well, I see you guys are talking about the indicators, some even based on the "sim/dis-sim zones" indicator to trade but as Eurusdd said it's just about 60% correct. Do you believe on something with 60% probability of correction or something which derives from that 60%? I'm not. So what I wonder is how to find "true" sim/dis-sim zones. That takes me come back to the fundamental concept: similarity. As Joe has just posted: {quote} So, What is similarity/dis-similarity? Why? When? How? I think if we can't find the clear and correct answers for...
okey, first, ci is like 100% correct with dissim thing
we do not need to know why dissim happens, we can use it anyway even if we don`t understand it
me thinks, that there is 2 kind of dissim things, but we don`t need what they are we can use them anyway

why Eurusdd is talking like he is talking, well, people has difficulties to keep secrets, because every one wants to be a hero, and in this case, all info is all ready given, he is trying to make chaos, because he feels bad, that he could not keep mouth shut, and trying to correct that mistake with chaos


one question: you can trade dissim with ci in m1 and m5 and get many signals every day, wtf is this stoch 300, why should you wait for weeks when you get many signals every day

fractal dimension with higher math is nothing else than dissim with fractal indi, normal thing for fractals is zigzag, dissim is there where this fails

this is my analysis about this thing
these levels was created in weekend from double fractals, eu-m5
Attached Image (click to enlarge)
Click to Enlarge Name: eurusdm5.png Size: 41 KB
  • #190
  • Jul 29, 2013 3:57am Jul 29, 2013 3:57am
Very interesting, do you think that the main function is that PRICE = STOCH?. This must be studied from all tf from highest to lowest?
  • #191
  • Jul 29, 2013 4:04am Jul 29, 2013 4:04am
it is a holy grail when all is put together, when you have everything in it
  • #192
  • Jul 29, 2013 4:10am Jul 29, 2013 4:10am
then the first part serious PRICE = STOCH to produce similarity
PRICE STOCH than would have dis-similarity.
This I would have to do for all tf.
If we PRICE = tf STOCH in 1day and
PRICE STOCH than in TF 4H, that means it is producing a change in trend, but lower tf guess the dis-similarity has already been produced not?. something more is needed to predict the turning point.
  • #193
  • Jul 29, 2013 4:10am Jul 29, 2013 4:10am
Quoting vlady1974
Some of you may know that I am trying to figure out a Daily Direction method..
Hi mate something I thought I would share with you....

If you look at the first three hours from london open you will see that most of the time the hi or low of the day is set....

So if we can identify either the hi or low of the day we can trade away either to set targets or ATR....

HOD= High of day

LOD= Low of the day
Attached Image (click to enlarge)
Click to Enlarge Name: lod.png Size: 30 KB
  • #194
  • Jul 29, 2013 4:15am Jul 29, 2013 4:15am
Quoting juhanimi
it is a holy grail when all is put together, when you have everything in it
Like EURUSDD said it took him 5 years to put all together.........a long time.
  • #195
  • Jul 29, 2013 4:16am Jul 29, 2013 4:16am
Quoting burnssss
Very interesting, do you think that the main function is that PRICE = STOCH?. This must be studied from all tf from highest to lowest?
For the BBZZ method yes. The Bollinger Band is set to 24 periods.
So the DS Time Unit is 300 and the Bollinger Time is 24

ie.
Price is Price
Stoch is 300 units of price (can someone mathematical or technical elaborate on this?) or setting of 300,1,1
Bollinger is 24.

I believe on the other thread, someone explained that the dissimilarity occurs when LL or HH enter the 300 time unit zone and basically disrupts the "averaged" result of the period. Continued trending results in DS until it stops. That DS "value" is then persisted?? over 24 units of time in a Bollinger Band to give you the extremes in that visual aid. Something like that.

So if you change the TIMEFRAME, you are in effect changing the number of bars you are representing on the MT4 screen.
My own questions:

  1. Is it possible to overlay all the timeframes together in one screen and then zoom out so you have 300 bars of the highest timeframe?
  2. 300 months is different to 300 weeks / days / etc. So what is the ideal or practical timeframe?
  3. 4500 minutes is the equivalent of the 300 x 15M unit that we all use. Translated, that is 3.125 days. Is this a magic number to anyone? Pi?

A pip is only worth it if you know how much you risked to earn it
  • #196
  • Jul 29, 2013 4:22am Jul 29, 2013 4:22am
Quoting mattkirby
{quote} Like EURUSDD said it took him 5 years to put all together.........a long time.
you just have to read what he writes, everything is there

question: how long have you been trading this?
EU: demo 2003, live 2004

he is 27, so... now is 2013, it was then years ago when he was demo trading, he was 17, so... it took 5 years to make this thing...
2003 - 5 years=1998-2000 or something...how old was he when he started to build this up.....12-13 years old, okey, maybe, or maybe not
  • #197
  • Jul 29, 2013 4:22am Jul 29, 2013 4:22am
Has anyone considered simple divergence when price and stoch are out of kilter?
  • #198
  • Jul 29, 2013 4:25am Jul 29, 2013 4:25am
Quoting juhanimi
{quote} you just have to read what he writes, everything is there question: how long have you been trading this? EU: demo 2003, live 2004 he is 27, so... now is 2013, it was then years ago when he was demo trading, he was 17, so... it took 5 years to make this thing... 2003 - 5 years=1998-2000 or something...how old was he when he started to build this up.....12-13 years old, okey, maybe, or maybe not
A pip is only worth it if you know how much you risked to earn it
  • #199
  • Jul 29, 2013 4:27am Jul 29, 2013 4:27am
On the 1H... we have a ZZTOP.
It's in the middle of the BB, so it's not valid. It's not resolved so we should expect a higher ZZTOP. (This is one of the main reasons I am calling for an UP DAY today - see other thread). Therefore by inference, all other lower timeframes needs to catch up to this dissimilarity.
A pip is only worth it if you know how much you risked to earn it
  • #200
  • Jul 29, 2013 4:29am Jul 29, 2013 4:29am
Good morning traders!

As far as my part in the experiment, I marked the DZ on H1

Attached Image (click to enlarge)
Click to Enlarge Name: usdchfh1.png Size: 46 KB


Then went down to M5 to find the trade, there was one when price was just above the zone with the following entry criteria met:

1. Price moved above band
2. zigzag
3. Semafor
4. Bearish price action candle.

Price then went on to take out this zone.

Attached Image (click to enlarge)
Click to Enlarge Name: usdchfm5.png Size: 46 KB


Perhaps this will spur further thought.
If you are good at something, never do it for free--Joker
3 traders viewing now