R - please share your experiences - page 7
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Please comment.
The model order is selected automatically using the Akaike information criterion. Read the help on the ar command.
Found
> x<-ar.burg(eur, aic=F, 20)
> x
Call:
ar.burg.default(x = eur, aic = F, order.max = 20)
Coefficients:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
0.9665 0.1096 -0.0941 0.0106 0.0004 0.0488 -0.0343 -0.0229 0.0288 0.0033 -0.0496 0.0168 0.0139 -0.0190 -0.0115 0.0173 0.0258 -0.0132 -0.0346 0.0365
As I understand it - it's a weighted mach in my example with T=20, only of higher quality. It differs only in the first term, which is a constant.
I wonder if it is possible to build TC on such scales?
As I understand it - this is the weighted waving in my example with T=20, only of higher quality. The only difference is the first term, which is a constant.
I wonder, is it possible to build TC on such scales?
You got it wrong, these models are not suitable for smoothing. Study the basics of econometrics.
Moreover, estimating AR models on data with a unit root will not lead to anything good.
You get it wrong, these models are not suitable for smoothing. Learn the basics of econometrics.
Moreover, estimating AR models on data with a unit root is not going to do any good.
Are you implying that you can't trust the coefficients because of MNC?
But there are a number of other estimation methods here, detrending....
Then, what's the stovepipe?
If econometrics, that's one thing, but if TA dummies, that's another. There's an estimation error here, and it's all darkness.By the way, I didn't copy it:
Order selected 20 sigma^2 estimated as 2.124e-06
There is an estimation error here, and there is solid darkness.
In your case, there is a model specification error.
In your case, there is a model specification error.
I understand that.
But what is the model specification error of a simple machine, an eXponential one, and where do we get the weighted coefficient to talk about errors? That's what I mean.
I understand that.
But what is the model specification error of a simple machine, an eXponential one, and where do we get the weighted coefficient to talk about the error? That's what I mean.
You don't understand. Learn the basics of econometrics. I'm not going to comment on further academic fluff.
The "mash-ups" have no specification error. Where to get weighted coefficients - study DSP.
You don't understand. Learn the basics of econometrics. I'm not going to comment on any further academic fluff.
There is no specification error in "wizards". Where to get weighted coefficients - study DSP.
Although harsh, thank you nonetheless.
I'll move on.
Help, problem again.
I evaluate regression: fm1 <- lm(dRegres1 ~ 1 + dRegres2, singular.ok = FALSE)
Everything is fine in R, but when I call it from mt4 I get an error:
Error in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) :
0 (non-NA) cases
What kills the most is that the debugged code in R doesn't work then in mt4.
Thanks in advance.