![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
For example, imagine a system of nested perfectly coinciding with the price digital filters with amplitude-frequency response of a damped oscillating link. For example, the first thing that catches your eye is coefficients that can be also compressed and stretched vertically without changing their shape, filters can be shifted to certain moments, and by predicting volatility we can predict this compression/stretch coefficient, which will set us boundaries of redrawing when shifting the filters.
I will find a couple of threads that I think will be useful for you in this regard.
https://forum.mql4.com/ru/45108/page41
https://forum.mql4.com/ru/45108/page44
AlexeyFX, most of his posts found in this forum can be read with interest but it seems to me that he lied about multicurrency.
Of course it's also interesting to predict their coefficients in order to predict non-harmonic vibrations. That is, if the market starts to flat, we decrease coefficients and make the filter lagging; if the situation is different, we increase coefficients and make the filter leading. Of course, it can be applied. But again, the problem is that the forecast is more or less accurate over a long period. I also thought about these methods, but I stumbled over this issue. Or should I redraw the filter for 12 hours ahead? Then the point is lost.
yours is not mura.... but it's secret...so there's no point in rubbing and flubbing. and repainted gates are often not recognised by intellectuals
Of course, it is also interesting with filters, to predict their coefficients in order to predict non-harmonic fluctuations. That is, if the market starts to flat, we decrease coefficients and make the filter lagging; if the situation is different, we increase coefficients and make the filter leading. Of course, it can be applied. But again, the problem is that the forecast is more or less accurate over a long period. I also thought about these methods, but I stumbled over this issue. Or should I redraw the filter for 12 hours ahead? Then the point is lost.
This one I don't quite understand, what do you take for x1, and for x2? Two wipers with different periods? And what are they? If I understand correctly, there will indeed be an oscillating line near 0, and you propose to normalize this oscillating line by a volatility prediction? Probably, if we do it right, we will get a prediction of the distance between these mashups, which we will use to determine when the trend is over. After all the fluctuations forecast is given for a long period, for example for 12 hours, that is, we know how much the price will pass in 12 hours (with a small error), if all these fluctuations are divided by each bar, we get a huge error. Or is it not necessary to divide by each bar, and only the total picture is enough?
x1 and x2 are for example the price of the first bar and the second one... I don't know about the rest, it probably depends on the algorithm of the forecast itself, I don't know how exactly you did it, there are different ways. Maybe you used tick volumes or equal-volume bars, or just time bars. And the time of analysis was used or not, maybe you built the distribution of bars by minutes within a day, that is, the series for the forecast were not a series of consecutive prices, and the high-low series of these values in a day for example.
here's an ostrich bending you over )))))
Cheeshim, there's an angry old man waiting for you in another thread, don't get distracted.
x1 and x2 are for example the price of the first bar and the second one... I don't understand the rest, it probably depends on the algorithm of the forecast itself, I don't know how exactly you did it, there are different ways. Maybe you used tick volumes or equal-volume bars, or just time bars. Yes and the time of analysis was used or not, maybe you build the distribution of bars by minutes within a day, that is, the series for the forecast were not a series of consecutive prices, and the high-low series of these values in a day for example.
Yes, the prediction is very simple, I have several methods there, but they all give almost the same result. I use linear prediction methods. I.e. I simply take ATR form for the previous period and use it to predict future values. I also tried Fourier forecast but it gives almost the same result but takes much longer.
I think it's nonsense to make separate forecasts for each timeframe.... , I should look for intertemporal links between the dynamics of changes in their tracks. Although we can go from the opposite direction, through the filters, making a difference between the price and the filter coefficients multiplied by the price, as in the example above.
Sneeze, there's an angry old geezer waiting for you in another thread, don't get distracted.
Ro-o-bot, ro-o-bot...
If you couldn't draw lines manually connecting price and grid intersection points with vertical step of 20...25 pips, then the robot probably won't help you...
However, good luck to you...
It is essentially the same as the renko bars only in the form of a line and what do you see as special about the renko there?
There's nothing in the Renko.
Are you not confused by the fact that the line is drawn - not the price...?