The similarity system | Page 555 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Similarity: A Practical Approach 859 replies

Similarity Trading - The ZZBB Method 215 replies

Rainbow similarity 256 replies

Similarity Stuff 17 replies

Directional Similarity 0 replies

Attachments: The similarity system
Exit Attachments

The similarity system

  • #11,081
  • Oct 27, 2014 11:29am Oct 27, 2014 11:29am
Quoting Rahimseven
For pondering and development. MM, trade entries, management and exits are your own secret. Decided to leave this nugget of info on the thread for those who may need a spark. This is but one understanding. Be careful as this is but a piece of the puzzle. Thanks for the kind PM's. 7 {image}
Thanks Rahim! Just read your post and skim through EURUSDD's posts again then something blows into my mind. I think I'm close to it... Now just need to work more on determining the best (h, k) and MM.

Guys, one of the key is here! Find the RARE EVENT!!!

Quoting Eurusdd
{quote} GOOD liquidgenius. Let me blow your mind now! If a bar has you rectangle then the probability the next one will have a rectangle is 0.00005463% on any time frame. NOW think what that means..... ONE THING: $$$$
  • #11,082
  • Oct 27, 2014 11:42am Oct 27, 2014 11:42am
Quoting Elhemano1
Ok, I have read the posts, but I have one question. When you got these "Zones" on above and below.. What to do and what does i means? {image}

Focus on the first zone. IMO

Also depends on how you would trade the zones
  • #11,083
  • Oct 27, 2014 12:39pm Oct 27, 2014 12:39pm
Some of the responses here are misleading! You have to pay a lot of attention to Definition.
The idea of a recurrent price is a global/universal concept with respect to a given probability measure. However, h-recurrent or h-transient is a local concept.
Therefore, please do not mix recurrent with h-recurrent.
If price is not hit-again with the next h-bars, price becomes h-transient. The story ends there. So, there is no need for people to talk about waiting forever or .....

It all depends on the h you set!

EVEN if price is hit after(not within) the next h-bars, it is STILL h-transient!!!!
  • #11,084
  • Oct 27, 2014 12:42pm Oct 27, 2014 12:42pm
Quoting Eurusdd
Some of the responses here are misleading! You have to pay a lot of attention to Definition. The idea of a recurrent price is a global/universal concept with respect to a given probability measure. However, h-recurrent or h-transient is a local concept. Therefore, please do not mix recurrent with h-recurrent. If price is not hit-again with the next h-bar, price becomes h-transient. The story ends there. So, there is no need for people to talk about waiting forever or ..... It all depends on the h you set!


hy eur/usd what i don't understand how do you know that a certain trade has a probability of over 90% to win?how to calculate that probability?and how do you do that in some post you say that if eur/usd doesn't reach a certain price in 2hours for example the probability is gone? i haven't found the answer anywhere
  • #11,085
  • Oct 27, 2014 12:43pm Oct 27, 2014 12:43pm
Quoting kore2
{quote} Hi Lak, I think it would be a better metaphor to say that you play your numbers on lottery week by week because you are 97% sure that those numbers would win. But you don`t know when. Maybe a year later. And you buy more than one lottery ticket. And you could go bankrupt while waiting for your lucky numbers.
Lack of understanding!!!!
  • #11,086
  • Oct 27, 2014 12:48pm Oct 27, 2014 12:48pm
This is my idea of transient/recurrent.
I still working in some ideas but i want that the recurrent price solve in "h" bars. I still working in this.
Than you very much
  • #11,087
  • Oct 27, 2014 12:51pm Oct 27, 2014 12:51pm
Quoting vladutz112
{quote} hy eur/usd what i don't understand how do you know that a certain trade has a probability of over 90% to win?how to calculate that probability?and how do you do that in some post you say that if eur/usd doesn't reach a certain price in 2hours for example the probability is gone? i haven't found the answer anywhere
Those probabilities are not true probabilities but estimators of the true probabilities, however with a large sample size and theorems from probability and Ergodic theories, there are way to be sure you are close to the true probabilities.
For example, recurrent theorems are ergodic theory are helpful. But you have to massage the data, something you can't do always.
  • #11,088
  • Oct 27, 2014 12:51pm Oct 27, 2014 12:51pm
Attached Image (click to enlarge)
Click to Enlarge Name: recurrent.jpg Size: 203 KB
Quoting Rahimseven
My ramblings. I know much is covered in the thread already but... 97% recurrent is a STATISTIC that MUST be qualified and understood for SELF before placing live funds into the game. Once understood, you can do the magic of forecasting probability and building equity with confidence. Testing this statistic is KEY! Input h,k, a data set and a time-frame using Eurusdd's trading revisits, over the past 100 days, hours, minutes ...Whatever Prove the recurrent principle for self confidence and responsibility. Take note of the attitude of price in and...
"Did price revisit the proposed recurrent area within h 97% of the time? yes?"

No, it didn`t. We can see 109 bars on this picture. We can see 3 completed transient zones (blue ones). So it is more than 97%. But pice did not revisit the proposed recurrent area 97% of the time. It did revisit it about 84% of the time.
It is easy to calculate it. We have 19 potential transient zones (salmon) on this picture. And 3 completed transient zones (blue). So what we have to calculate is 19:3 (instead of 109:3).
So it`s not 97% of the time but only about 84% of the time.

Btw thanks a lot for your pdfs.
  • #11,089
  • Oct 27, 2014 12:54pm Oct 27, 2014 12:54pm
Quoting Eurusdd
{quote} Lack of understanding!!!!
You are right. I am still searching and searching.
  • #11,090
  • Oct 27, 2014 12:55pm Oct 27, 2014 12:55pm
Quoting kore2
{image}{quote} "Did price revisit the proposed recurrent area within h 97% of the time? yes?" No, it didn`t. We can see 109 bars on this picture. We can see 3 completed transient zones (blue ones). So it is more than 97%. But pice did not revisit the proposed recurrent area 97% of the time. It did revisit it about 84% of the time. It is easy to calculate it. We have 19 potential transient zones (salmon) on this picture. And 3 completed transient zones (blue). So what we have to calculate is 19:3 (instead of 109:3). So it`s not 97% of the time but...
Your understanding of the whole concept is classically flawed. In fact, put the concept aside, your probability skills are bad. How can you use a few sample points of your chart to infer any sensible truth?

You need to understand h-recurrent prices. Then you need to understand what almost-surely means. then you need to understand how to calculate probabilities correctly
  • #11,091
  • Oct 27, 2014 1:02pm Oct 27, 2014 1:02pm
Quoting kore2
{quote} You are right. I am still searching and searching.
I tossed a coin and i got -H-H--H-H-H-H-H-H-H-H-H-H-H-H-H-H-H--T-T--H-H-H-H-H-H-H-H.


Those who said T must appear half of the time, are wrong! Is that a sound statement?.


If you have not got this yet, then you cannot follow what will come later.
  • #11,092
  • Oct 27, 2014 1:06pm Oct 27, 2014 1:06pm
Quoting Eurusdd
{quote} Your understanding of the whole concept is classically flawed. In fact, put the concept aside, your probability skills are bad. How can you use a few sample points of your chart to infer any sensible truth? You need to understand h-recurrent prices. Then you need to understand what almost-surely means. then you need to understand how to calculate probabilities correctly
You said:
"Therefore, please do not mix recurrent with h-recurrent."

That`s what I wanted to differentiate on that picture.
So did I mess it up?
  • #11,093
  • Oct 27, 2014 1:09pm Oct 27, 2014 1:09pm
Quoting kore2
{quote} You said: "Therefore, please do not mix recurrent with h-recurrent." That`s what I wanted to differentiate on that picture. So did I mess it up?
He was referring to earlier post
  • #11,094
  • Oct 27, 2014 1:16pm Oct 27, 2014 1:16pm
Quoting LITEchild
{quote} You are still posting rubbish as usual I can see. Buisness must be picking up. Posting account statements with an average of 3pips per trade??? What a joke! And completely contradicting everything about TZs and probabilities. I hope other less discerning traders will take the time to analyse your trades entries and exits and see that you obvious don't know what you're doing. Your account statement says 18 trades have been closed. You showed us 16! What were the losses like on the 2 trades you're hiding??? Look at the trading account statement...
hello Lite, i Tried to look but i cant see the TE's you talked about, can someone help please ?
  • #11,095
  • Oct 27, 2014 1:20pm Oct 27, 2014 1:20pm
The proof of proposition 1 is very technical and I do not believe it is important for us, unless you want to understand the math. What is important is this: the statement is independent of the PAIR you trade. Once the process is governed by human-emotions and bounded is a certain sense, the proposition will hold!
Therefore, as more and more data .... in, we should get closer and closer to the truth!

If you think it is wrong, disprove the proposition!
Applying a theorem is a different matter all together.
  • #11,096
  • Oct 27, 2014 1:23pm Oct 27, 2014 1:23pm
Quoting Eurusdd
{quote} I tossed a coin and i got -H-H--H-H-H-H-H-H-H-H-H-H-H-H-H-H-H--T-T--H-H-H-H-H-H-H-H. Those who said T must appear half of the time, are wrong! Is that a sound statement? ...
You tossed 20+ times here. But I guess if you toss 200, 2000 or 20000 times ... the chances a coin toss end up head or tail ... should approach 50-50?
  • #11,097
  • Oct 27, 2014 1:24pm Oct 27, 2014 1:24pm
Maybe k should be fibonacci numbers...
No matter how much you know, your words worth only as much as the other per
  • #11,098
  • Oct 27, 2014 1:27pm Oct 27, 2014 1:27pm
Quoting Eurusdd
{quote} Your understanding of the whole concept is classically flawed. In fact, put the concept aside, your probability skills are bad. How can you use a few sample points of your chart to infer any sensible truth? You need to understand h-recurrent prices. Then you need to understand what almost-surely means. then you need to understand how to calculate probabilities correctly
did sunsaltwater understood correctly your concept? cuz he's the one that gave clear indication an the only one that made me understood what it's about the tz zones and you also said in one post that his explanations made your day but others seems that see him as a fraud
  • #11,099
  • Oct 27, 2014 1:27pm Oct 27, 2014 1:27pm
Quoting summicron
{quote} You tossed 20+ times here. But I guess if you toss 200, 2000 or 20000 times ... the chances a coin toss end up head or tail ... should approach 50-50?
I used that example, to show the flaw in his argument. He used a few sample points on his chart to say 84% is more like it and not 97%+.

That is why I said, you cannot say 50:50 is wrong from that 20 tosses.
  • #11,100
  • Oct 27, 2014 1:28pm Oct 27, 2014 1:28pm
Quoting vladutz112
{quote} did sunsaltwater understood correctly your concept? cuz he's the one that gave clear indication an the only one that made me understood what it's about the tz zones and you also said in one post that his explanations made your day but others seems that see him as a fraud
I am not following this thread always, so maybe I missed his other posts. The one I read though was good!
15 traders viewing now