a trading strategy based on Elliott Wave Theory - page 90

New comment
 
It doesn't work....Then follow the link.
 
Rosh, don't you like the solution? "If there's no difference, why pay more?" :))
And Solandra (thanks to him for his participation in this thread) has killed 630 lines in this case ..... with the same accuracy...
By the way - if someone calculates other quantiles I will help with formulas.
 
2 Yrixx,
I'm very ashamed of course, but I'll ask anyway. How did you get this formula D(E) = D(Y) - a^2*D(X). If you can do it for the same price, please sell the deduction logic.) Or give me a hint where to read it.
 
I'm very embarrassed, but I'll ask anyway. How did you get this formula D(E) = D(Y) - a^2*D(X). If you can at the same price sell the deduction logic ;). Or give me a hint where to read it.

No problem to your email. I gave mine in this thread.
 
<br / translate="no"> No problem to your email. I gave mine in this thread.


Searched and searched, couldn't find it. Here's mine ********
 
Looking at this picture


a thought occurred to me that maybe this notorious minimum of the functional is inherent to the channel whose parabola coefficient A->0. I.e. the field sources on top of the regression line and on the bottom counterbalance each other.

And here is the situation, the channels I have chosen have the following RMS

2006.07.17 20:11:30 VGGopII EURUSD,H4: N= 160 CKO2/3= 0.00707805 CKO= 0.00739682
2006.07.17 20:11:30 VGGopII EURUSD,H4: N= 307 CKO2/3= 0.00863145 CKO= 0.00967016

We see that these channels have CKO>CO2/3. You can argue about the smaller one, but the bigger one can be seen even by eye as a channel.
 
Looking at this picture a thought occurred to me that maybe this notorious minimum of the functional is inherent to the channel with the parabola coefficient A->0. I.e. the field sources on top of the regression line and on the bottom counterbalance each other. <br/ translate="no">.


Correction. The sign of parameter A for a parabola doesn't matter, you have to relate it to the A sign of the linear regression (from general considerations). This is not because of my secret knowledge, but for the love of art :)

ZS. Did you teach the parabola to draw outside the sample boundaries? Congratulations, I can't get my hands on it.
 
<br / translate="no">

ZS. Did you teach the parabola to draw beyond the sampling boundaries? Congratulations, I can't get my hands on it.

I just looked at how ANG3110 did it in his script.
if (i<0) { datetime te=Time[0]-(i)*kt; datetime te1=Time[0]-(i+1)*kt; ObjectMove(n+"par "+i,0,te1,fx1); ObjectMove(n+"par "+i,1,te,fx); ObjectSet(n+"par "+i,OBJPROP_COLOR,Lime); ObjectSet(n+"par "+i,OBJPROP_WIDTH,3); } else { ObjectMove(n+"par "+i,1,Time[i+1],fx1); ObjectMove(n+"par "+i,0,Time[i],fx); }


 
Rosh, don't you like the solution? "If there's no difference, why pay more?" :)) <br / translate="no"> And Solandra (thanks to him for participation in this thread) killed 630 lines in this case... ..... with the same accuracy...
By the way - if someone calculates other quantiles I will help with formulas.


But this is a private solution, and I like general ones. Or am I missing something?
I did not see the expression of probability through the random deviation from the regression centre.
1...838485868788899091929394959697...309
New comment