a trading strategy based on Elliott Wave Theory - page 79

New comment
 
Yes, I too, as Solandr not able to understand how you have Rosh occur such fast calculations, of course I have done through a bunch of functions, but get rid of them I do not want to, as with the algorithm is not all well will be to catch problem areas, I'm probably premenyu improved method of skipping bars for small channels, it will go straight and with increasing period increases the number of skipped bars.

At the sequential analysis of each channel my algorithm has found channels with a period of 144, 1044, 2689 and calculation is by eye about 13 sec (on Celeron 2.4) P = 14.79%
with 5 bars skipping -- 145, 1045, 2690 and it takes about 4 sec P=14.82%
with 20 bars -- 160, 1080, 2700 and it took about 1 second P=15.16%

so I see that the margin of error is acceptable and I'm going to move in that direction for now.
 
Solandr, if it's not too much trouble for you, could you make a caption with the length of the channel?

Why are you so sensitive to those channels my Expert Advisor is looking for? I have already told you that Vladislav searches for them using something else, i.e., I search them using the minimum RMS and it does not mean that they are absolutely correct. As an option for channel search I gave my variant of search by sum of gradients. I haven't implemented that yet. I think that plus or minus some error in determining the channels themselves can only affect about the same percentage of profit. So, it is more important to test the Expert Advisor rather than endlessly catching bars. If your Expert Advisor will confidently lose, because you haven't yet defined the criteria for opening/closing positions, then what difference does it make if the channels diverge by 1-2% in length? Why do you need absolutely accurate channels without a strategy itself? Do you want to trade manually? I already know for sure that manual trading is very stressful and time consuming. If you want to trade manually, you may manually adjust the channels by moving their beginnings to the extremes without any field theories - like all traders do.
 
This utility is not working correctly for me. I have uploaded the missing files to the system directory.
I have all the files in the same folder, haven't copied anything anywhere.
Just checked, everything works.
I have WinXP Pro SP2 ru
 
Solandr если Вас это не затруднит, не моглибы Вы псделать подписи с длинной канала?

Why are you so squeamish about the channels that my EA is looking for? I already told you that Vladislav searches for them using something else, i.e. I search for them using the minimum RMS and it does not mean that they are absolutely correct. I gave my variant of search by sum of gradients as an option for searching channels. I haven't implemented that yet. I think that plus or minus some error in determining the channels themselves can only affect about the same percentage of profit. So, it is more important to test the Expert Advisor rather than endlessly catching bars. If your Expert Advisor will confidently lose, because you haven't yet defined the criteria for opening/closing positions, then what difference does it make if the channels diverge by 1-2% in length? Why do you need absolutely accurate channels without a strategy itself? Do you want to trade manually? I already know for sure that manual trading is very stressful and time consuming. If you want to trade manually, you may manually adjust the channels by moving their beginnings to the extremes without any field theories - like all traders do.



I think the picture was posted to let other comers compare your results, because in my opinion you are the most advanced in building a strategy. So trying to compare it with yours, it was enough to see the whole charts or know their periods, or to be more precise, I was interested only in the biggest one, because it mainly defines the probability, at least for me. Of course the channel with long period of 100 days cannot fit on the chart, that is why I wanted to ask you for it. And the difference of 10-15% from your drawing would allow me to assume that I am moving in the right direction and have not missed any important details. No, you're not. :(

I agree with you about manual trading, and I am trying my best to get away from it too.
 
Yes, I'm the same as Solandr can not understand how you Rosh make such fast calculations, of course I have done everything through a bunch of functions, but I do not want to get rid of them yet because of the algorithm is not all good will catch the problem areas, I probably will apply the improved method of skipping bars, for small channels, it will go straight and with increasing period to increase the number of skipped bars.
I see that the error is acceptable and will still move in that direction



I don't know, I've gone the opposite way. I debug all necessary functions at once and collect them in the Channels library (the source code size is 32 kb) with maximum documentedness. Then scripts, indicators simply include this library through #include.
The size of the source code, which draws a lot of channels from any point on the history, is 5400 bytes, the size of indicator, which is now 15 pairs, is 6100 bytes, and the EA will also be small. I still need to add an option to select prices, timeframe, number of channels to display and calculate Ttudent (I have already loaded the normal distribution as an array).
But sometimes I am too lazy to correct mistakes and instead of doing it correctly I use z-z-z-z. As a result, I've paid for it today - the terminal doesn't slow down, but it eats memory to the maximum.
I'll have to set the array's size not to Bars (because of laziness), but to a proper value - to the sampling depth.

 
I understand that the picture was posted by you so that other participants could compare their results with yours, because in my opinion you are the most advanced in building a strategy. So trying to compare it with yours, it was enough to see the whole charts or know their periods, or to be more precise, I was interested only in the biggest one, because it mainly defines the probability, at least for me. Of course the channel with long period of 100 days cannot fit on the chart, that is why I wanted to ask you for it. And the difference of 10-15% from your drawing would allow me to assume that I am moving in the right direction and have not missed any important details. No, you're not. :(

Okay. It's no problem for me to punch in the bar number in the comments. Here's a link to pictures, look, compare, if it can really help https://c.mql5.com/mql4/forum/2006/07/10_07_2006_additional.zip
And so in my opinion everyone is moving in the same right direction. It's just really hard enough to go in different directions implementing the same rule.
 
<br / translate="no"> OK. It's no problem for me to put the bar number in the comments. Here's a link to the pictures, have a look, compare, if it can really help https://c.mql5.com/mql4/forum/2006/07/10_07_2006_additional.zip
And so in my opinion everyone is moving in the same right direction. It's just really hard enough to go in different directions implementing the same rule.


Not a problem for me either :)



и

 
solandr Question for you as the most advanced member of the sect.
Here is the picture of 15 min Eur , MMLevls_VG with MMPeriod=60. As far as I understand (I haven't studied Murray yet) - time to buy with a small stop m target at 1.2787/1.2817/1.2848 ?

 
I have all the files in the same folder, haven't copied anything anywhere. <br / translate="no"> Just checked, everything works.
I have WinXP Pro SP2 ru

I have Windows2000SP4 (freshly installed on a separate machine with nothing but the operating system and the downloaded utility itself and a file to edit). Is it really necessary to intentionally install WinXP? If so, in principle, it is not a problem to experiment with WinXP (I restore it from an image).
 
As far as I understand (I haven't studied Murray yet) - time to buy with a small stop and target at 1.2787/1.2817/1.2848 ?

I think it is too early to stockpile the euro. So far the channel pattern seems to be in favour of the Euro. Why do you think about buying the Euro, if you have all the signs that it is going down? In fact, in my pictures the average probability of going up/down is around 50%. My expert is waiting it out for now. No open positions.
1...727374757677787980818283848586...309
New comment
Reason: