The similarity system | Page 550 | 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

  • #10,981
  • Edited 6:33am Oct 25, 2014 5:04am | Edited 6:33am
EDITED.......GOT THE SOURCE
re:i just posted this on other thread,
I HAD this on my laptop from somewhere i cant remember and am sharing it ,may be it will be useful for those that want to analyse transient zones with volume. I wont say more than that.
Pulse and think mostly on eurusdd way of thinking on price behavior.

How the price makes and how the price up and down? I’ll give you example on “How the price make?” at a simplified simulation.

Simulation of: “HOW PRICE CHANGE?” in simplified.
Name of multi-Currency: GBPJPY Present price: 172.354 bid, 172.379 ask (example only)
Name of Institution that look into this price: MM1, MM2, MM3.
MM1 already hold their position open in market as long. They have 20 open long already. Here is the list of their offer (bid) price:
172.600 (2 Position)
172.750 (3 Position)
173.100 (4 position)
173.300 (11 position)
MM2 and the other looking for Short position, so, they look to Short at highest price as possible.
What happened next when some of the MM fill or do a deal in market?
MM2 Open short by fill order MM1 at 172.600. The price jump up from 172.354 bid into 172.600 bid
and 172.620 ask. This order makes the new price now updated into 172.600 (bid) and 172.620 (ask).
This is the new price.
Then MM3 Open short by fill order MM1 at 172.750. The price jump up from 172.600 bid into
172.750 bid and 172.780 ask. This order makes the new price now updated into 172.750 (bid) and
172.780 (ask). This is the new price.

The prices itself already SHOW the position that MM look for in the future. As you see from example above, price jump up when the MM opening short. Jump up again, when another MM opening short again. I ask you to keep this as a secret (Ha ha ha ha….). So, no wonder if retail trader 90% suffered loss. The trend that many retail trader follow, actually is the result of the opposite position from MM. They know the future, but retail trader not. Then, if we are in the same page here, we can
conclude with: “price trend up, we looking for short; price trend down, we looking for long.”
By understand “price up, we looking for short; price down, we looking for long.”, doesn’t mean that
we should blindly jump in, and open position.
Attached File(s)
File Type: pdf Exploit the market price by Volume reading.pdf   827 KB | 1,775 downloads
We Win Some, We Loose Some, But Let your Win Rates be More to be Successful
  • #10,982
  • Oct 25, 2014 5:19am Oct 25, 2014 5:19am
Quoting Kiads
{quote} Thank You so much My Friend..... {image} best regards MTH
Beautiful! Your charts are like eye candy! I could stare at them all day ...and forget to trade
  • #10,983
  • Oct 25, 2014 5:34am Oct 25, 2014 5:34am
Quoting Sciurus
Someone asked for a version of my recurrence indy having the option to specify left and right hand h separately. Here it is. {file}
Very kind of you buddy. Thank you.
  • #10,984
  • Oct 25, 2014 6:06am Oct 25, 2014 6:06am
Had a eureka moment as I created this for my notes. Maybe you will have one as well.

Enjoy...
Attached File(s)
File Type: pdf Eurusdd's TZ RZ Clues and Lessons.pdf   823 KB | 2,789 downloads
  • #10,985
  • Oct 25, 2014 8:39am Oct 25, 2014 8:39am
Quoting Kiads
{quote} And speaking of divergences, maybe my funny indicator can help a little bit.. or a little bad.. lol MTH Edit : How to... ?? {image} Enjoy {file}
Dear Kiads

Will you please share these indis and this template in a zip file format?


Regards
Traders gain some knowledge, win, their emotions kick in, and they die.
  • #10,986
  • Oct 25, 2014 8:51am Oct 25, 2014 8:51am
Quoting Rahimseven
Had a eureka moment as I created this for my notes. Maybe you will have one as well. Enjoy... {image}
Good work! Thank you.
I've read at least 5 times those posts from EURUSDD, yet haven't figured out some specific things.

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% (later corrected 5.463%) on any time frame. NOW think what that means..... ONE THING: $$$$
We know that the current bar has a rectangle (transient zone) only after h bars passed, so what next bar is EURUSDD referring to? Next bar after h bars completed the TZ? Or completely wrong understanding?

Quoting Eurusdd
So, if you find one instance on your chart where your choice of h or k did not work or price did not behave as you expected immediately, that is a rare event. There next instance is most likely to work So, do not raise your hand to the sky cursing the powers that be. Have more confidence that you will not be living to see a rare-rare event. Probability is on your side. The next instance is good to go...
Hmm... Has this todo with mid-bar transient zones? Rare-Rare event?


Somehow I missed the following post (http://www.forexfactory.com/showthread.php?p=7681717#post7681717):
Quoting Eurusdd
{quote} This is important. For example, if your k is constant, i.e. k=5, and there is a pin-bar with tail = 10 pips, you are supposed to declare only 5 pips of whatever part is left as transient in your trading process. Don't extend the transient portion to the whole tail. so even though the light-blue part was not really recurrent, your sell still worked!!! {image} .
K variable exposed
EURUSDD said once that k should be small, less than 10% of "active trading window" (something similar). Then he mentioned his sure 5-10 pips trading the 1min and 5min TF. Now open your charts and look on the 5min TF, measure pips from top to bottom of the chart window, most likely result 10 pips (or 5 pips on 1min tf etc...).

I'll try to study and integrate this k thing now in my TZ trading.
the 3rd wave is the crowd's wave
  • #10,987
  • Oct 25, 2014 9:03am Oct 25, 2014 9:03am
"We know that the current bar has a rectangle (transient zone) only after h bars passed, so what next bar is EURUSDD referring to? Next bar after h bars completed the TZ? Or completely wrong understanding?"

according to EURUSDD probability calculation only %5 of chance price will not touch that level & there will be another rectangle "partial TZ" will be created.
so what would you do?
%95 price will go through
%5 it will stay away

GL

Quoting juriFX
{quote} Good work! Thank you. I've read at least 5 times those posts from EURUSDD, yet haven't figured out some specific things. {quote} We know that the current bar has a rectangle (transient zone) only after h bars passed, so what next bar is EURUSDD referring to? Next bar after h bars completed the TZ? Or completely wrong understanding? {quote} Hmm... Has this todo with mid-bar transient zones? Rare-Rare event? Somehow I missed the following post (http://www.forexfactory.com/showthread.php?p=7681717#post7681717): {quote} K variable...
My Best Trade is Next Trade!
  • #10,988
  • Oct 25, 2014 9:25am Oct 25, 2014 9:25am
" Somehow I missed the following post (http://www.forexfactory.com/showthread.php?p=7681717#post7681717): {quote} K variable exposed EURUSDD said once that k should be small, less than 10% of "active trading window" (something similar). Then he mentioned his sure 5-10 pips trading the 1min and 5min TF. Now open your charts and look on the 5min TF, measure pips from top to bottom of the chart window, most likely result 10 pips (or 5 pips on 1min tf etc...). I'll try to study and integrate this k thing now in my TZ trading."

That is indeed a great post IMO. Remember "How you calculate k is up to you. That will differentiate you from another person applying the same principle. Keep that secret." Eurusdd...
  • #10,989
  • Oct 25, 2014 9:29am Oct 25, 2014 9:29am
So just to be clear. K is NOT the width of the TZ...
  • #10,990
  • Oct 25, 2014 9:40am Oct 25, 2014 9:40am
Quoting Sciurus
Someone asked for a version of my recurrence indy having the option to specify left and right hand h separately. Here it is. {file}
thanks very very nice
regards orid peter
  • #10,991
  • Oct 25, 2014 9:42am Oct 25, 2014 9:42am
Quoting Eurusdd
{quote} This is important. For example, if your k is constant, i.e. k=5, and there is a pin-bar with tail = 10 pips, you are supposed to declare only 5 pips of whatever part is left as transient in your trading process. Don't extend the transient portion to the whole tail. so even though the light-blue part was not really recurrent, your sell still worked!!! {image} .

Major jewels right here.
Attached Image (click to enlarge)
Click to Enlarge Name: Capture.PNG Size: 33 KB
  • #10,992
  • Edited 11:30am Oct 25, 2014 11:18am | Edited 11:30am
OK, since the topic of k is being discussed, I think it timely to give further description of how and why I coded my indy the way I did. My original idea was to not only discover the best value for h, but also for k. My vision of k is the same as Rahimseven's. Good for him to pick that out from EURUSD's comments, when it would be easy to overlook it as a typo.
Below you will see two screenshots. In the first, I have attached my indicator to a 5 minute chart. You will notice that for lookback bars I have entered a very large number, this forces the indy to default to all bars in history. You will also notice two numbers in the upper portion of the screen. The first is a probability figure (in this case 0.088). I chose to express probability as a value from 0 to 1. This is normal in probability theory. It can be translated as 8.8%, hence what this means is that 8.8% of potentially transient price historically became confirmed transient. notice that I do not say bars - I say price. The calculation is based upon the total height of the boxes you see on the screen. Some bars are bigger than others, and some potential transient zones are bigger than others.

Next you will see the upper 95% confidence interval being equal to 7.8 pips. This is an analysis of the zones that became confirmed transient. So I took the average height in pips of all the transient zones, and then I added 1.96 times the standard error of all of those zones. This is a quick and dirty way of saying that 95% of all transient zones have a height below 7.8 pips. Ok, maybe we are starting to see where this is going...
** important** the 95% CI calculation does not take into account the zones that were cleared! Hence, it is an overestimation! But I prefer to overestimate in this case until i know what I am dealing with. Hey, here is an idea, lets use 7.8 as my value for k! cool, but how can we visualize k on the chart??

Now look at my next screenshot. I have entered the value of 8 (I could have used 7.8, but I rounded) in the minPipWidthBack parameter. This will exclude all potential transient zones less than 8 pips in height. The 95% CI value now becomes a bit meaningless. the probability is still a bit interesting. It actually went up a bit. Perhaps the probability of a transient zone being cleared is not totally independent from potential zone height to bar size. Only a further in depth statistical analysis would reveal this. But notice what happened. now there are many fewer zones being displayed. It is only showing the zones greater than our assumed k = 95% CI. I have put approximate white lines where price passed this value for k within each zone. note that even for the one zone that eventually became transient, price passed this value for k and then returned. **Note, within the framework I have just divulged, these white lines would be TARGETS, not ENTRIES.

Ok. please do not misunderstand what I have just said. Do your due diligence, and don't take what I have just discussed as a means to make instant money. It is meant to facilitate a process of discovery.

Another thing, Bonus points to anyone that can figure out what the channel setting is. That isn't supposed to be there, sorry.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: indy1.jpg Size: 440 KB
Click to Enlarge Name: indy2.jpg Size: 463 KB
  • #10,993
  • Edited 11:45am Oct 25, 2014 11:23am | Edited 11:45am
This thread just keeps on giving when reading past posts.

Combining some cool thing at the moment. I'll share the ideas of the concept after testing!

Peace

Ullen
Attached Image (click to enlarge)
Click to Enlarge Name: Skærmbillede 2014-10-25 kl. 18.21.00.png Size: 25 KB
Not about how much you win, but how much you loose
  • #10,994
  • Oct 25, 2014 11:46am Oct 25, 2014 11:46am
Quoting Sciurus
OK, since the topic of k is being discussed, I think it timely to give further description of how and why I coded my indy the way I did. My original idea was to not only discover the best value for h, but also for k. My vision of k is the same as Rahimseven's. Good for him to pick that out from EURUSD's comments, when it would be easy to overlook it as a typo
Hi, if I understand this, you calculate k based on the width of true transient zones.
While that is perfectly fine, and also my indicator does that, this is exactly what Rahim says that is not correct, and he is right. k from Proposition 1 is linked with the recurrent interval (it is half of its width), not the transient one.
What I (and I believe you do the same) call k in my indicator is a useful quantity, but is just bad nomenclature to call it k. I am equally guilty of this.

k
  • #10,995
  • Oct 25, 2014 11:46am Oct 25, 2014 11:46am
Quoting Sciurus
OK, since the topic of k is being discussed, I think it timely to give further description of how and why I coded my indy the way I did. My original idea was to not only discover the best value for h, but also for k. My vision of k is the same as Rahimseven's. Good for him to pick that out from EURUSD's comments, when it would be easy to overlook it as a typo. Below you will see two screenshots. In the first, I have attached my indicator to a 5 minute chart. You will notice that for lookback bars I have entered a very large number, this forces the...
Wow! Great post!
  • #10,996
  • Oct 25, 2014 11:50am Oct 25, 2014 11:50am
Quoting MAM.Trader
according to EURUSDD probability calculation only %5 of chance price will not touch that level & there will be another rectangle "partial TZ" will be created. so what would you do? %95 price will go through %5 it will stay away GL {quote}
Sorry still don't understand.

Original statement:
Quoting Eurusdd
{quote} If a bar has you rectangle then the probability the next one will have a rectangle is 5% on any time frame. NOW think what that means..... ONE THING: $$$$
So let me break that down so hopefully someone can spot the loophole in my reasoning.

1. "a bar has you rectangle": A bar gets a rectangle if it becomes a TZ, so h bars passed. Maybe EURUSDD is referring to a potential TZ forming? And the rectangle bar would be the central bar of the PTZ...

2. "the probability the next one will have a rectangle is 5%": If in 1. a confirmed TZ is meant, then the next bar would be h_right+1. If in 1. a PTZ bar is meant, then I can't see how the next bar has only 5% probability of invalidating the PTZ.

This chart is a good example, my statistics say price has around 33% probability to invalidate the PTZ on the very next candle bar.
Attached Image
the 3rd wave is the crowd's wave
  • #10,997
  • Oct 25, 2014 11:58am Oct 25, 2014 11:58am
if you wanna bring us somewhere by asking pls dont
say it

you say
"my statistics say price has around 33% probability to invalidate the PTZ on the very next candle bar"

eurusdd say
"my statistics say price has around 95% probability to invalidate the PTZ on the very next candle bar"


Quoting juriFX
{quote} Sorry still don't understand. Original statement: {quote} So let me break that down so hopefully someone can spot the loophole in my reasoning. 1. "a bar has you rectangle": A bar gets a rectangle if it becomes a TZ, so h bars passed. Maybe EURUSDD is referring to a potential TZ forming? And the rectangle bar would be the central bar of the PTZ... 2. "the probability the next one will have a rectangle is 5%": If in 1. a confirmed TZ is meant, then the next bar would be h_right+1. If in 1. a PTZ bar is meant, then I can't see how the next...
My Best Trade is Next Trade!
  • #10,998
  • Edited 12:19pm Oct 25, 2014 12:09pm | Edited 12:19pm
Quoting juriFX
{quote} Sorry still don't understand. Original statement: {quote} So let me break that down so hopefully someone can spot the loophole in my reasoning. 1. "a bar has you rectangle": A bar gets a rectangle if it becomes a TZ, so h bars passed. Maybe EURUSDD is referring to a potential TZ forming? And the rectangle bar would be the central bar of the PTZ... 2. "the probability the next one will have a rectangle is 5%": If in 1. a confirmed TZ is meant, then the next bar would be h_right+1. If in 1. a PTZ bar is meant, then I can't see how the next...
From my understanding so far:

Eurusdd refered to a PTZ bar where contains current transient prices status. According to his stats, 95% chance this transient prices would not hold (to become a confirmed TZ (5%), that means this transient status would change to recurrent status (95%) - and this is where money is - and we work out a strategy to exploit this. Yours is 33% that is due to your h value.

If I am wrong please correct me buddies.

Regards,
Exciter135.
  • #10,999
  • Oct 25, 2014 12:20pm Oct 25, 2014 12:20pm
Quoting MAM.Trader
if you wanna bring us somewhere by asking pls dont say it you say "my statistics say price has around 33% probability to invalidate the PTZ on the very next candle bar" eurusdd say "my statistics say price has around 95% probability to invalidate the PTZ on the very next candle bar" {quote}
Quoting Exciter135
{quote} From my understanding so far: Eurusdd refered to a PTZ bar where contains current transient prices status. According to his stats, 95% chance this transient prices would not hold (to become a confirmed TZ (5%), that means this transient status would change to recurrent status (95%) - and this is where money is - and we work out a strategy to exploit this. Yours is 33% that is due to your h value. If I am wrong please correct me buddies. Regards, Exciter135.
OK now all clear. I thought EURUSDD was saying the opposite hence my lack of understanding. Thanks.

Quoting Sciurus
OK, since the topic of k is being discussed, I think it timely to give further description of how and why I coded my indy the way I did. My original idea was to not only discover the best value for h, but also for k. My vision of k is the same as Rahimseven's. Good for him to pick that out from EURUSD's comments, when it would be easy to overlook it as a typo. Below you will see two screenshots. In the first, I have attached my indicator to a 5 minute chart. You will notice that for lookback bars I have entered a very large number, this forces the...
Good indicator and good insights! Right on time for new experiments with k
Thanks for sharing.

Quoting Ullen
This thread just keeps on giving when reading past posts. Combining some cool thing at the moment. I'll share the ideas of the concept after testing! Peace Ullen {image}
Looking forward it.


Quoting kprsa
{quote} Hi, if I understand this, you calculate k based on the width of true transient zones. While that is perfectly fine, and also my indicator does that, this is exactly what Rahim says that is not correct, and he is right. k from Proposition 1 is linked with the recurrent interval (it is half of its width), not the transient one. What I (and I believe you do the same) call k in my indicator is a useful quantity, but is just bad nomenclature to call it k. I am equally guilty of this. k
Proposition 1 seems in conflict with the example setup (http://www.forexfactory.com/showthread.php?p=7681717#post7681717) EURUSDD showed us, where he clearly stated k is the height of the transient zone.
the 3rd wave is the crowd's wave
  • #11,000
  • Oct 25, 2014 12:25pm Oct 25, 2014 12:25pm
[quote=Sciurus;7824077]. please do not misunderstand what I have just said. Do your due diligence, and don't take what I have just discussed as a means to make instant money. It is meant to facilitate a process of discovery. quote]

Ok assuming I understood you right, then using your indicator and referring to these "sample" figures you used.

1)When I see a salmon box that box I'm looking at has less than 10% chance to become transient so I'm expecting ~90% probability that the "Painting" blue box is going to disappear? Or does the blue box only get painted when a TZ is confirmed?

2)So I could in theory...also target the blue boxes aside from the white lines in the salmon box? Sorry the white lines are not showing up on my chart so I can't really tell if I'm talking about the same thing

3)The salmon doesn't go away after price has eaten them right?
I want more brain power!
12 traders viewing now