I made one of these things once ... - 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
у меня вообще с этим индикатором фантастика какая то, ты да и многие на форуме знают, что я с ним бодался очень долго и упорно, хотите верте хотите нет, но расскажу ибо до сих пор сам в шоковом состоянии от этого …
Did I understand you correctly that it took you many weeks of rigorous calculations in higher mathematics (integration, differentiation, approximation, etc., etc.) to understand that even financial markets are seasonal and tend to turn around at the beginning of the month? Yet you still have no idea how to use it?
.... By the way, my opinion, the pullback from 1.27 was supported by the Swiss, besides the fact that it was a strong round level, it was the first one to rise on the news, and then the Euro went up. The CHF is very strong there, if the CHF breaks through 1.05 and the EUR 1.27 (which I think it will) the move will be really tough...
....
https://www.mql5.com/ru/forum/126769/page99#345261
I missed it, I wrote about it, the level of 1.28 does not change the essence of the method, it's just a question of our mathematical researches ))
Did I understand you correctly that it took you many weeks of rigorous calculations in higher mathematics (integration, differentiation, approximation, etc., etc.) to understand that even financial markets have seasonality and tend to turn around at the beginning of the month? Yet you still have no idea how to use it?
I understand how to use it, I hope you don't think I'm a circular .... There's a forecast thread with my use of it (feel free to double-check it). Another thing that struck me, I didn't program it, didn't expect it at all...how should I put it...let's say you make a dummy, and you know it lags...and then you open the chart and see there is no lag, not at all, you just need to look at it from a slightly different angle (I came across it by accident), not the way you planned to use it...it is kind of a by-product...
Still, it's better to use the visualiser first :)
No problem pulling it out, it is of course specific information, but if you need it you are welcome to do so
If you look carefully at these charts (links above) you can see that the reversal, occurs at the beginning of the month (EXACTLY at the beginning!!!), and it has no input parameters, the only thing I can change is the start point (from where it should be calculated on the history) and that's it...
The only thing I can change is the start point (from which it should be calculated on historical data) and that's all. I couldn't get past that fact, I haven't programmed it, it doesn't even know if it was a month, week or year...
I know this algorithm too well. And it's not a matter of wrong hands. It needs to be controlled, which is easy to do visually, but I have no idea how to do it programmatically.
I'm with my favourite grid.
I have done such a thing, wrote down the number of points inside the figure for the zigzag vertices and plotted the distribution. It looks like this
We can see that at levels x.xx00 the tops of the zigzag are really more attractive, but at levels x.xx50 no features are visible.
No, I didn't leave Kalman. It was the one I was struggling with the longest. I brought it to the necessary level in MQL , although I could go on and on, but it started working the way I wanted it to, that's enough for me.
Here it is, this is what I was struggling with it for
https://www.mql5.com/ru/forum/126769/page92#345010
https://www.mql5.com/ru/forum/126769/page72
....
Lately it has become fashionable to blame everything on the context, and it seems to rule here.
...
We can see that at x.xx00 levels the zigzag tops do like to be located a bit more, but at x.xx50 levels no specifics are visible.
i don't need to cheat anyone, why ? well, look how you like it, the indicator started, counted, counted, and stopped
I can see it visually. I tried to use your algorithm to build a fast zigzag. There is a restart if it fails, but it may not start at all, like a motorbike, you yank the button until it growls, roars, roars and stops. If you look closely, you can see the thin blue stripes, which means it is not 100% right, its readings should also be interpreted ...
as for the zigzagging levels, you can, but look at
For me it's important to enter the market. Almost immediately the market may be lossless, but if it reaches 1.29 it's a different matter.
At the same time, the zigzag will not break at these points, though it has exact stops at the level.
like the dollar franc today
but notice again there is an entry with a minimum stop...
Pretty accurate predictions of the beginnings of a reversal. Have you tried Particle Filters where, unlike Kalman filters, the noise statistics are estimated from the data itself instead of the Gaussian distribution hypothesis? I'm currently trying to master these partial filters in matlab.
strange this is the second time i've seen this talk about Gaussian distribution of noise. strange. can i get a source for that?
strange this is the second time I see that they talk about Gaussian noise distribution. strange. where is it mentioned?
Wikipedia, Kalman filter:
The Kalman filter model assumes the true state at time k is evolved from the state at(k - 1) according to
where
At time k an observation (or measurement) zk of the true state xk is made according to
where Hk is the observation model which maps the true state space into the observed space and vk is the observation noise which is assumed to be zero mean Gaussian white noise with covariance Rk.
see how you like it, the indicator starts, counts, counts, and stops
as for the zigzagging levels, you can, but look
...at the same time, as I understand it, the zigzag will not have a break exactly at these points, near, yes, at the same time there are really stops exactly at the level.