Statistics as a way of looking into the future! - page 9
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
Well, I don't know how else to explain it!!! Think a little bit about what you're posting.
I wrote that it's not Close :-), it's an estimate. That it should ideally be like this (if the model was consistent with the process). Give me some time to lay out the files and you'll understand what I build :-)
Well, Prival gave the correct tangent, but he calculated it relative to the predicted and estimated curve. The problem is that the estimated curve has too little in common with the real price to use it. Which is what I've shown in the previous charts.
Yeah. We'll trade on the real curve and show pictures on the estimated curve! Is that how it works?
I wrote that it's not Close :-) but estimate. I said it should ideally be like that (if the model corresponds to the process). Give a little time I will lay out the files and you'll understand that from what I was building :-)
Well, it's done!
I told you just above that it will soon become clear why Prival is not in a good mood and I won't trash my NS developments :-)))
bstone писал(а) >>
Ну Prival правильный тангенс выдал, но считал он его относительно прогнозной и оценочной кривой.Проблема же в том, что оценочная кривая имеет слишком мало общего с реальной ценой, чтобы это можно было использовать. Что я и показал на предыдущих графиках.
that's right, it's the same but in different words. If the model would match, it's not even a chocolate, it's already a grail :-). But the indicator is interesting, it lets you get something, not much, but a bit.
Here is a picture I like it better :-) allows to bite
Yeah. We'll trade on the real curve and show pictures on the estimated curve! Is that how it works?
Well as far as I understand, Prival is working on forecasting exactly the estimated curve. Therefore the result is very good for him. The practical implications remain a mystery.
If the model matched, it's not even chocolate, it's already a grail :-).
Almost fooled me :-)
Almost fooled me :-)
I didn't do it on purpose, here's the file, see what's built from what. I was trying to tell you, you can see it's sloppy. You need a more accurate model, then it will work. Although thoughts and ideas how to use it would be interesting.
Well as far as I understand, Prival is working on forecasting exactly the estimation curve. Therefore the result is very good for him. The practical meaning is still a mystery.
Look at the image above, the red curve has very good properties, it's smooth (I can adjust it) and has less lag (I can also adjust it) comparing to other price indicators I know.
At the bottom is an oscillator based on an estimate and forecast.
Neutron, what is the tangent of your NS-workings on H1?
I work with NS on binary inputs i.e. with signs of price increments. There are several reasons for that. And the main one is the learning curve. Therefore, I cannot build a forecast cloud, because it is built for real (or integer) numbers. But I can calculate the share of correct entries. My proportion is 30-50% (depending on time horizon), where 0% corresponds to "random entries" - 50/50 and 100% - "all guessed".