Testing real-time forecasting systems - page 29
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
Today was a perfect day's work
I have this situation:
in theory, the process should unfold ("reflect"). We'll see.
The tweaking consisted of removing pattern price smoothing and increasing sigma (reducing cluster size) so that the closest patterns affected the prediction.
Thank you.
I've been working with NS for a long time, but these nets have passed my way, I've only glanced at them in general outline. So I have had an idea, to improve my statistics by these nets (others objectively do not suit me very much). As free time, will try to implement. In this regard, can you be pestered with questions?
Grasn, if I understand correctly the forecast is the centre line and the upper and lower are the confidence intervals? It would be a good idea to make different coloured lines, because it all gets mixed up.
On the whole, we are in the same direction, which makes me happy :-)
Grasn, if I understand correctly the forecast is the centre line and the upper and lower are the confidence intervals? It would be a good idea to make different coloured lines, because everything gets mixed up.
>> On the whole we coincide, what can't make me happier :-)
The calculation in Matcad, the result was on page 25, here it is
:
I see it better here. And the MT screenshots are my mastery of MQL (between you and me, I think the prefies are biting their elbows with envy). The dark grey boundary is 1 RMS built from the MO process. By and large, it's rare for a process implementation to fit within such a "narrow" boundary. The worm inside, is a refinement of the process.
It is not a trajectory as such in the sense of a prediction, it is the areas where increased concentration of the quotation process is to be expected. What is pictured is a first approximation, the process is iterative. In addition, alternative realisations are considered.
Thank you.
I've been dealing with NS for a long time, but these networks have gone by the wayside, I've only glimpsed them in general terms. I have got an idea to improve my statistics by means of these nets (other ones are not really suitable). As free time, will try to implement. In this connection, it will be possible to you popstavit with questions?
It's a good time to get a little bit of a head start.
I see that we all have similar predictions: EURUSD will trade sideways and then up.
GRNN:
There's plenty of room for molestation.
I see that we have all formed similar predictions: EURUSD will trade sideways and then up.
Ah, what - we can co-operate, make up a macro-strategy - have a TC poll, vote and select a candidate. :о)))))
I think it's reflected, the areas are a bit displaced of course, but that's OK, I'll work on it and see what happens next ...
Last year's prediction by the GRNN network was pretty accurate:
Here's the new prediction
Last year's prediction by the GRNN network was pretty accurate:
Here's a new prediction.
Yes, it's a beauty.
My extrapolator did pretty well too, here's the new one: