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.
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
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.
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.
Allow the use of cookies to log in to the MQL5.com website.
Please enable the necessary setting in your browser, otherwise you will not be able to log in.
You are signed in. Two-factor authentication has been disabled. Set it up again and scan a new QR code, or continue with two-factor authentication disabled.
[img]https://c.mql5.com/mql4/forum/2006/07/%D1%F2%FC%FE%E4%E5%ED%F21.gif [/img]
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.
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.
No problem to your email. I gave mine in this thread.
Searched and searched, couldn't find it. Here's mine ********
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.
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.
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); }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.