{quote} Nah.. you just need this..{image}
No matter how much you know, your words worth only as much as the other per
- #11,141
- Oct 28, 2014 2:34pm Oct 28, 2014 2:34pm
- #11,142
- Oct 28, 2014 3:26pm Oct 28, 2014 3:26pm
- | Membership Revoked | Joined Oct 2012 | 2,254 Posts
{quote} Kiads, we see that you have this picture. Please do not share Picture of your equipments. Shortly (k)iads+ use it on your (t)ransient+(h)oles ahh sorry (z)ones... YES I found formula k+t+h+z
- #11,145
- Edited 4:13pm Oct 28, 2014 3:44pm | Edited 4:13pm
I coded RevealWormhole for 4 different timeframe pairs: 1 minute and 5 minutes, 5 minute and 15 minutes, 15 minutes and 30 minutes, 30 minutes and 1 hour. You can control the K period values with external settings. The defaults are not necessarily the best so you can easily change them per your discretion. I'll post the EA source code later.
Open to new approaches.
- #11,146
- Oct 28, 2014 4:51pm Oct 28, 2014 4:51pm
- Joined Nov 2013 | Status: Member of the 5% club | 1,259 Posts
{quote} Here they are. My observation: 1min_5min generates a lot of noise. Maybe someone will figure out how to use that one to their advantage. {file} {file} {file} {file}
{quote}Hey Saver0, Do you mean youre working with dynamic h values (i.e. different h values on the same time frame depending on price action) or its just the way you calculate probabilities thats different? Because its not quite clear how the zone you highlighted here (http://www.forexfactory.com/showthread.php?p=7829615#post7829615) would not be an h-recurrent one. I think this is what kprsa was trying to say. Kind Regards.
For the price at current bar at this moment to be recurrent or transient, doesn't matter, what conditional shall H bars ago must satisfy?
When you answer that question correctly, you will understand what I was saying.
In my example, I demonstrated a zone or price level(s) in blue that is not Recurrent or Transient with a H setting of 80 (80 is the width of the zone in bars).
Green pips everybody!
Focus, Patience, Determination & Order in chaos
- #11,149
- Oct 28, 2014 7:22pm Oct 28, 2014 7:22pm
{quote} Let me try to answer your question with another question. For the price at current bar at this moment to be recurrent or transient, doesn't matter, what conditional shall H bars ago must satisfy?(*1) When you answer that question correctly, you will understand what I was saying. In my example, I demonstrated a zone or price level(s) in blue that is not Recurrent or Transient with a H setting of 80 (*2) (80 is the width of the zone in bars). Green pips everybody!
The statement at *2 might be contentious for this reason: the price at a current moment temporarily qualifies as Transient if it has not been printed on the chart within the last H bars into the past, and is confirmed as Transient if it is not printed on the chart in the next future H bars. Isnt this what you understand by EURUSDD's proposition? I think all the prices in the zone you highlighted are printed at least once again on the chart within 80 bars into the past or 80 bars into the future from each price in that zone? (which, I'd suppose, would make them all '80-recurrent')
- #11,152
- Oct 28, 2014 8:13pm Oct 28, 2014 8:13pm
- | Membership Revoked | Joined May 2012 | 1,218 Posts
{quote} Eurusdd, first of all if i am dude, you are dude too. Second, thank you for your works here(i am extremely respectfull and appreciated but i do not care). Third, Look at up, what kiads( a kind of noob) sent me...
and create a nice..
and think for a while what the meaning of that....
until you can understand and see similarity on this....
or if you still don't get it you can ask for...
or if you don't like the result you can say..
what a wonderful world..
MTH
- #11,154
- Oct 28, 2014 8:39pm Oct 28, 2014 8:39pm
{quote} Lol.. it's best tools for you.., you can make up your mouth but more important make up your mind {image} and create a nice.. {image} and think for a while what the meaning of that.... {image} until you can understand and see similarity on this.... {image} or if you still don't get it you can ask for... {image} or if you don't like the result you can say.. {image} what a wonderful world.. MTH
. The more I see from you the more I like you. 
- #11,155
- Oct 28, 2014 9:25pm Oct 28, 2014 9:25pm
- Joined May 2013 | Status: Digging deeper | 653 Posts
Hello everyone,
There seems to be a lot of debate and confusion about probabilities at the moment, both on this thread and vlady's thread. I admit that I don't have much background in this subject, and by the looks of it I'm not the only one. I think it would be really helpful to us all if someone who has a deep understanding of probability could explain, as simply as possible, which assumption is right and which one isn't. Or are they both right? I know it might not really matter, so long as our trades work, and this isn't a classroom and no one's obliged to teach us anything. But I think it would benefit the thread a lot.
So here goes. This is my question:
1. Assume we have 1000 bars of data (this is just an example, I know sample size should be much larger than this).
2. Assume that h=5, where h is the number of bars before and after the middle bar. So total width of h-zone is 11 (including the middle bar).
3. Assume that within these 1000 bars, 970 are h-recurrent. That is, the entire price range of the middle bar is "touched " by at least one other bar within the h-zone.
4. Therefore, 30 bars are h-transient. That is, one or more prices within the price range of the middle bar are not "touched" by any of the other bars within the h-zone.
5. Therefore, 97% bars of the sample belong to the h-recurrent group, and 3% of the bars belong to the h-transient group.
6. Therefore, the empirical probability of any bar being h-recurrent is 97%.
So far so good, right? I think most all of us will agree with this up to now.
7. Next, let's split the h-recurrent group unto 3 sub-groups:
- i) h-left-recurrent (middle bar prices only touched by bar(s) to the left)
- ii) h-right-recurrent (middle bar prices only touched by bar(s) to the right)
- iii) h-fully-recurrent (middle bar prices touched by bars to the left AND right)
8. Let's assume that the bar counts of these sub-groups are as follows:
- i) h-left-recurrent = 180 bars
- ii) h-right-recurrent = 110 bars
- iii) h-fully-recurrent = 680 bars
9. Therefore, the empirical probability of any bar belonging to the following groups is:
- i) h-left-recurrent = 18%
- ii) h-right-recurrent = 11%
- iii) h-fully-recurrent = 68%
- iv) h-transient = 3%
10. This gives us a total of 100%, as expected.
So far so good? I'm assuming it's OK to include the h-transient group with the other three, even though it's not a sub-group?
So, all this is leading up to my big question:
11. Assume that you are looking at a middle bar and you can't see what's to the right of it (like in real trading).
12. Assume that the middle bar is h-left-transient (some prices not touched within h bars to the left).
13. Therefore, after h bars to the right, this middle bar will belong to one of two groups:
- i) h-right-recurrent, OR
- ii) h-transient
(it's impossible for it to belong to the h-left-recurrent or h-fully-recurrent groups)
Finally, assume that you want to calculate the probability of this middle bar becoming h-right-recurrent. What do you do? I think there are two options:
A) Since the total number of bars it could turn into is 140 (110 h-right-recurrent + 30 h-transient), then the probability is 110 / 140 = 78.5%, OR
B) Since it will also turn into either the larger h-recurrent group (970 bars) or the h-transient group (30 bars), then the probability is 970 / 1000 = 97%.
This is my dilema. Which calculation is correct? They can't both be correct, can they? Or do you take an average of the two?
I'm inclined to think that (A) makes more sense logically, but I could be wrong of course.
Thanks for reading this and sorry if it isn't clear. I'm looking forward to the explanation.
There seems to be a lot of debate and confusion about probabilities at the moment, both on this thread and vlady's thread. I admit that I don't have much background in this subject, and by the looks of it I'm not the only one. I think it would be really helpful to us all if someone who has a deep understanding of probability could explain, as simply as possible, which assumption is right and which one isn't. Or are they both right? I know it might not really matter, so long as our trades work, and this isn't a classroom and no one's obliged to teach us anything. But I think it would benefit the thread a lot.
So here goes. This is my question:
1. Assume we have 1000 bars of data (this is just an example, I know sample size should be much larger than this).
2. Assume that h=5, where h is the number of bars before and after the middle bar. So total width of h-zone is 11 (including the middle bar).
3. Assume that within these 1000 bars, 970 are h-recurrent. That is, the entire price range of the middle bar is "touched " by at least one other bar within the h-zone.
4. Therefore, 30 bars are h-transient. That is, one or more prices within the price range of the middle bar are not "touched" by any of the other bars within the h-zone.
5. Therefore, 97% bars of the sample belong to the h-recurrent group, and 3% of the bars belong to the h-transient group.
6. Therefore, the empirical probability of any bar being h-recurrent is 97%.
So far so good, right? I think most all of us will agree with this up to now.
7. Next, let's split the h-recurrent group unto 3 sub-groups:
- i) h-left-recurrent (middle bar prices only touched by bar(s) to the left)
- ii) h-right-recurrent (middle bar prices only touched by bar(s) to the right)
- iii) h-fully-recurrent (middle bar prices touched by bars to the left AND right)
8. Let's assume that the bar counts of these sub-groups are as follows:
- i) h-left-recurrent = 180 bars
- ii) h-right-recurrent = 110 bars
- iii) h-fully-recurrent = 680 bars
9. Therefore, the empirical probability of any bar belonging to the following groups is:
- i) h-left-recurrent = 18%
- ii) h-right-recurrent = 11%
- iii) h-fully-recurrent = 68%
- iv) h-transient = 3%
10. This gives us a total of 100%, as expected.
So far so good? I'm assuming it's OK to include the h-transient group with the other three, even though it's not a sub-group?
So, all this is leading up to my big question:
11. Assume that you are looking at a middle bar and you can't see what's to the right of it (like in real trading).
12. Assume that the middle bar is h-left-transient (some prices not touched within h bars to the left).
13. Therefore, after h bars to the right, this middle bar will belong to one of two groups:
- i) h-right-recurrent, OR
- ii) h-transient
(it's impossible for it to belong to the h-left-recurrent or h-fully-recurrent groups)
Finally, assume that you want to calculate the probability of this middle bar becoming h-right-recurrent. What do you do? I think there are two options:
A) Since the total number of bars it could turn into is 140 (110 h-right-recurrent + 30 h-transient), then the probability is 110 / 140 = 78.5%, OR
B) Since it will also turn into either the larger h-recurrent group (970 bars) or the h-transient group (30 bars), then the probability is 970 / 1000 = 97%.
This is my dilema. Which calculation is correct? They can't both be correct, can they? Or do you take an average of the two?
I'm inclined to think that (A) makes more sense logically, but I could be wrong of course.
Thanks for reading this and sorry if it isn't clear. I'm looking forward to the explanation.
You reap what you sow.
- #11,156
- Oct 28, 2014 11:06pm Oct 28, 2014 11:06pm
- | Additional Username | Joined May 2013 | 724 Posts
Hello everyone, There seems to be a lot of debate and confusion about probabilities at the moment, both on this thread and vlady's thread. I admit that I don't have much background in this subject, and by the looks of it I'm not the only one. I think it would be really helpful to us all if someone who has a deep understanding of probability could explain, as simply as possible, which assumption is right and which one isn't. Or are they both right? I know it might not really matter, so long as our trades work, and this isn't a classroom and no one's...
This is how you ask a question properly!
I want more brain power!
- #11,157
- Oct 28, 2014 11:21pm Oct 28, 2014 11:21pm
- | Membership Revoked | Joined May 2012 | 1,218 Posts
Anyone attempting to trade into potential transient zones without a solid framework should become familiar with the zigzag indicator, because that would basically be what you would be doing. Try this, attach the zigzag indicator and let its period = h. Then compare where the zigzag draws its legs to where the confirmed zones are. In theory, the zigzag will start a new leg whenever a new potential zone appears, and repaint whenever it is cleared to a lower low/higher high.
- #11,158
- Oct 28, 2014 11:26pm Oct 28, 2014 11:26pm
{quote} Lol.. it's best tools for you.., you can make up your mouth but more important make up your mind {image} and create a nice.. {image} and think for a while what the meaning of that.... {image} until you can understand and see similarity on this.... {image} or if you still don't get it you can ask for... {image} or if you don't like the result you can say.. {image} what a wonderful world.. MTH
Send more, i will help you!
No matter how much you know, your words worth only as much as the other per
