From theory to practice - page 189

 
bas:

Have you tried the HP filter? I think there was an induced one in the kodobase.

I don't need one. I count the filters I need on my own.

bas:

p.s. Polynomial is wrong at reversals (as far as I remember my experiments from 5 years ago), because its ends fly to the sky, which is unusual for market. And it does not try to give an average, but adjusts to the whole quote.

You must be confused about something. About sky-high - that's more to do with FFT. I mentioned polynomial regression only as an example, but quite possible.

bas:

By the way, even 50 pages ago I've shown, that your "weights for WMA" have no sense, as the result doesn't differ a bit from SMA )))

If it's really so, it's even worse than expected.( Systematic error in distribution calculation is off-scale. To the eye, somewhere around +/-30% on average in each count. Well, let's say even 20%.

 
Yuriy Asaulenko If indeed this is the case, it's even worse than expected.

This is because the author does not think about the physical meaning of his actions - he takes the increments as weights. If he had taken deviations as weights, the result would be more meaningful (known as weighted approximation, lowess or something like that).

 
bas:

This is because the author does not think about the physical meaning of his actions - he takes the increments as weights. If he had taken deviations as weights, the result would be more meaningful (known as weighted approximation, lowess or something like that).

No matter how you take it, the error in the mean will not go anywhere.

 
Yuriy Asaulenko You must be confused about something. About skyward - that's more to do with FFT. I mentioned polynomial regression only as an example, but quite possible.

Well, for example, a quadratic - it tends far from the centre:


 
bas:

For example, a quadratic - it tends to go way off-centre:

Imho, it's quite all right.) It looks quite consistent. Why does it have to be in the centre? It shouldn't, and you can see that.

Never mind the regression. That's not what we're talking about.)

 
bas:

Well for example a quadratic - it tends to go far off-centre:


is it redrawn?

========

PS

just to think...

a correct indicator should not redraw by 1 point

Forecast on H4 - 4 hours ahead

MN forecast - 1 month ahead

Good luck!

 
bas:

For example, a quadratic - it tends to go way off-centre:


and you listen more to the bicycle masters giving out free advice


 
Maxim Dmitrievsky:

and you listen more to bike lords giving out free advice


Why is this Jurik not Jurik?

Why is polynom not mashka?

Why is this drawing genius

makes us sell poop?)))

 
Vizard_:

Why is this Jurik not Jurik?

Why isn't Polynom a mashka?

Why is this drawing genius

makes us sell poo?)))

How come, brothers?

♪ We've learned the polynomials in full force ♪

♪ We keep on grieving ♪

♪ And all our sorrow is from our minds ♪

 
Maxim Dmitrievsky:

How is it possible, brothers?

♪ We know the polynomials in full measure ♪

We keep on grieving

♪ And all the grief we feel is from our minds ♪

¶¶ There's something wrong with the polynomials ¶¶

And there's something wrong in our heads

Don't tell Yurik

He's not Yura, he's Vanya the Fool)))

Reason: