Econometrics: one step ahead forecast - page 80
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
to faa1947
I don't really understand why you're showing off. Do you think you are the only one fighting against TA and VA? No, ask my colleagues and they will not lie - I am the most vicious opponent of these so-called directions. I quarrel and fight with everyone here :o). But everyone chooses his own path, yours is a strange one, to put it mildly, and you can see it very well, because it's mathematics at least.
Yes, I can't stand "econometricians" (I often have to deal with them at main work), mainly because they simply don't understand what they do, they take from various fields what's bad and literally stupidly start using, without understanding what they hold in their hands.
This:
EURUSD = -1552.7613734*DXM_HP(-1) + 4731.89082764*DXM_HP(-2) - 4360.68995095*DXM_HP(-3) + 1287.82064375*DXM_HP(-4) - 98.9244837504*DXM_HP_D(-1) - 131.011472103*DXM_HP_D(-2)
HP is a piece of it.
Name me any indicator in the code base for which you know the R-square
is complete nonsense. The R-square has nothing to do with it and in this case it is of no practical use and shows absolute rubbish. Is it not clear? Don't you understand that you can't use it? Here is a 5000 bars long kotir, M15 EUR and its ACF (as a source for 500 oscillations):
Here's an ACF like this will give you any perfect linear model. For comparison, here is a perfectly random series (you can model it as you wish) of the same length 5000
And here is its ACF (original source, without increments):
This ACF will also give you any perfect linear model. And you think you can make money on it? How much? It's like a very simple example, very simple and that time has been wasted on you. Well, maybe you'll finally start to understand what you're teaching. And my advice to you - stop showing off, it causes nothing but irritation.
PS: If by some miracle you get a statistical advantage, then do not rejoice... Ask the market about your econometrics, which by the way you have not written anything about.
>>>I'm sorry if that was harsh - I do have a grudge against those econometricians :o)))
Interesting, but out-of-sample, i.e. on a real road. And what econometrics offers is just a fitting for a vehicle - a muzzle, which looks like a real car in the showroom (on the optimization sample), but when you go to the highway (out of sample) turns out to be a trough unsuitable for driving - the wheels fall off, instead of the engine bricks.
Well look at the tables: out-of-sample prediction, out-of-sample prediction, out-of-sample prediction.
to faa1947
I don't really understand why you're showing off. Do you think you are the only one fighting against TA and VA? No, ask my colleagues and they will not lie - I am the most vicious opponent of these so-called directions. I quarrel and fight with everyone here :o). But everyone chooses his own path, yours is a strange one, to put it mildly, and you can see it very well, because it's mathematics at least.
Yes, I can't stand "econometricians" (I often have to deal with them at main work), mainly because they simply don't understand what they do, they take from various fields what's bad and literally stupidly start using, without understanding what they hold in their hands.
This:
is complete nonsense. The R square has nothing to do with it, and in this case it has no practical use and shows absolute nonsense. Is it not clear? Don't you understand that you can't use it? Here is a 5000 bars long kotir, M15 EUR and its ACF (as a source for 500 oscillations):
Here's an ACF like this will give you any perfect linear model. For comparison, here is a perfectly random series (you can model it as you wish) of the same length 5000
And here is its ACF (original source, without increments):
This ACF will also give you any perfect linear model. And you think you can make money on it? How much? It's like a very simple example, very simple and that time has been wasted on you. Well, maybe you'll finally start to understand what you're teaching. And my advice to you - stop showing off, it causes nothing but irritation.
PS: If by some miracle you get a statistical advantage, then do not rejoice... Ask the market about your econometrics, which by the way you have not written anything about.
>>>I'm sorry if that was harsh - I do have a grudge against those econometricians :o)))
Yes, it's time to wind down. Even literate people don't bother to write anything of substance. Attributed something to me, got indignant. Carry on, but without me.
Yes, it's time to wind down. Even literate people don't bother to write anything of substance. Attributed something to me, got indignant. Carry on, but without me.
Yes, it's time to wrap it up.
[...]Carry on, but without me.
No, you can't bail. It's your karma, you can't escape it.
If you leave, nothing will change, and the same karma will still have to be worked out elsewhere. You still do not understand the reasons why such a beautiful model does not work.
....
AutoCorrelationFunction() - is this a built-in function or self-written?
Sorry, I don't have the matcad at hand to check it.
If the sample is 500 samples, the ACF should be zero at 500 shifts. This is where I once posted the ACF.
https://www.mql5.com/ru/code/8295
If you need (Skype privalov-sv) I'll look for my calculations in matcad. for comparison. there are 3 different ways to calculate...
P/S/ What do the abbreviations "You alone are struggling with TA and VA" mean.
Thanks.
Working TS has a profit factor of one and a half at most. And if it's 5 - then you can throw it away immediately without optimisation.
hehe :)))) Paukas, I'm doing an experiment with the whole year, and now the second month is over and the profit-factor = empty, i.e. infinity :D)))
Should I throw it away?
It's all the same: traders are only interested in the right side of the yield curve - OOS, econometricians are only interested in the left side - the fit.
The point is that neither traders nor econometrists care about the results that are on the part they are not interested in. Therefore, traders and econometricians have nothing in common.
How many deals in those two months?