how to identify price reversal points - page 16
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
With the right input, the output is not important.
and vice versa.
I tried to make my own SHI_Channel_true indicator based on your indicator to calculate the slope angle and width of the channel. I have come to the conclusion that the indicator does not work correctly. It works by fractals in the interval from 0 to AllBars (after modification - from Start to Start+AllBars). So, it works correctly if the beginning or the end of the interval falls on a price reversal. Otherwise the channel direction starts to change almost like the МАшашка - now it goes up and now it goes down. The conclusion is that we should use fractals tied to the turning point of the price. In this case, it is better to determine them by a zigzag (breakpoints coincide with the support-resistance levels- this is what is needed for calculation of the channel). To sum it up: we need a channel indicator built on the basis of fractals built on the basis of a zigzag. I could certainly do it, but I would lose a lot of time. Maybe you have seen such indicators?
I tried to make my own SHI_Channel_true indicator based on your indicator to calculate the slope angle and width of the channel. I have come to the conclusion that the indicator does not work correctly. It works by fractals in the interval from 0 to AllBars (after modification - from Start to Start+AllBars). So, it works correctly if the beginning or the end of the interval falls on a price reversal. Otherwise the channel direction starts to vary almost like the МАSHA - now it goes up and now it goes down. The conclusion is that we should use fractals tied to the turning point of the price. In this case, it is better to determine them by a zigzag (breakpoints coincide with the support-resistance levels - this is what is needed for calculation of the channel). To sum it up: we need a channel indicator built on the basis of fractals built on the basis of a zigzag. I could certainly do it, but I would lose a lot of time. Maybe you have seen such indicators?
I read it, it's interesting,
I hope no one will dispute that the correctness of an indicator signal based on history is 50/50 )))
The task is very simple
- to identify the local price reversal points. In fact, it is the potential entry points into the market (they just need to be filtered by the rules of TS).
So, I cannot find a suitable indicator that would help us clearly identify such points. The figure shows the adaptive fractal average in blue, the channel zigzag in purple and linearly smoothed 1st derivative of the fractal average (i.e., the rate of change of the fractal average) in green. Vertical lines at the intersection with the blue and green chart form extreme points of price fluctuations (these are almost entry points).
The chart becomes jerky - there are a lot of false signals. When smoothing it lags a lot. I cannot find the "golden mean" when the points do not lag too much and there are not so many false signals. Do you think some other indicators may be applied to determine the entry points?
Let's start with the definitions (as Reshetov says - the correct wording is almost a solution) of what you want to identify:
- local pivot points are ... (the correct wording of what you mean by this) (bar closing points or?) (the reversal is counted by the number of bars, by time, by target, ...?)
- potential market entry points are ... (and all local pivot points)
- if they need to be filtered, what are the rules?
- clearly define - it means ... (within a bar, by its close, by bar extremums, ...) (which bar ? current ? previous ?...)
- extreme points of price fluctuations, on which intervals?
- what does the notion of a false signal depend on, how is it formulated?
- what do points not lag too much ? (how much is it?)
- False signals are few, i.e. how many?
as for the indicators - just indicators that detect three-bar fractals are enough - i.e. you identify all extrema - the only thing left is to filter out the unsuitable ones (i.e. whether the signals of these 3-bar fractals produce exit points that correspond to your objectives).
...points in time corresponding to extreme values of the averaged price in a given time interval. The law of averaging is not necessarily linear...
...points of time, when an order opened in a specified direction is likely to close in profit.
I repeatedly wrote above. In brief - the general rules of trend and channel trading.
To have the least amount of false points with the minimum delay.
Defined parametrically.
The absence of false signals - is when the real signal does not change at the same moment in time when the signal is calculated again (in history).
It means the lag time is minimal. I consider 1-3 bars on H1 interval as an acceptable lag.
After the full filtration of signals by TS, no more than 10-15% of false signals should remain.
Produce after filtering.
And-and-and... now are you going to tell me some interesting solution?
...Time points corresponding to extremes of the averaged price in a given time interval. The law of averaging is not necessarily linear...
- the word - reversal - is lost here
...points in time, when an order opened in a given direction is likely to close with a profit.
- Assuming the direction is correct, then we are talking about the exit point (closing the order)
I have written above on several occasions. In brief, the general rules for trend and channel trading.
- Is the channel sloping or horizontal?
To have as few false points as possible with minimal delay.
- we do not know if this point is false or not on the current bar
It is set parametrically.
- Ok
The absence of false signals is when the real signal does not change when the signal is recalculated (already in history) at the same point in time.
- ... but more ...
So the lag time is minimal. I consider 1-3 bars allowable lag time on H1 interval.
- Okay .
After the full filtration of signals by TS, no more than 10-15% of false signals should remain.
- good
Produce after filtering.
And-and-and... now you will tell me some interesting solution?
- I will, but on condition of barter exchange (I don't know how interesting it is for you) if you have something interesting for me from your developments
adept:
...time moments corresponding to extreme values of the averaged price in a given time interval. The law of averaging is not necessarily linear...
- there's a word lost here - U-turn
The reversal is at the moment of extreme value. At least that is what I mean.
adept:
...points in time when an order opened in a given direction is likely to close with a profit.
- If the direction is correctly determined, we are speaking about the exit point (order closing)
And so is the exit. Sometimes. This is a separate task. For the time being we are talking about the entrance. Entry at the moment of extreme value. The direction of the reversal determines the direction of entry.
adept:
I repeatedly wrote above. In brief, the general rules for trend and channel trading.
- Is the channel inclined or horizontal?
Either one. In the first case, the entry direction coincides with the trend direction, in the second case entry depends on the channel width - the trend is not important.
adept:
To have as few false points as possible with minimal delay.
- we do not know if this point is false on the current bar
Yes. Validation can only be done on historical data. We cannot look into the future.
adept:
The absence of false signals is when the real signal does not change when the signal is recalculated (already in history) at the same moment in time.
- but more details...
At time T0 we calculate the signal and get S0. At time T1 we calculate the signal at time T0 and get S1. No false signals is when for any T0 and T1 at T1 > T0, S0 = S1.
adept:
- I will give you a hint, but on condition of barter exchange (I do not know how interesting it is for you), if you have something from your developments that is of equivalent interest to meWhat exactly do you offer? Can write in person ...
The reversal is at the point of extremum. At least that is what I mean.
- a reversal has a price or % value, for me it is a path of more than 30 pips after the extremum, and for you? And we will never know if it is 30 pips until it is over, but even 5 pips is a reversal for the pipser, so it makes no sense to define a reversal without exit points
And so is the exit. Sometimes. This is a separate task. For the time being we are talking about the entrance. Entry at the moment of extreme value. The direction of the reversal determines the direction of entry.
...
Yes. Validation can only be done on historical data. We cannot look into the future.
- Again, whether it is a false point or not depends on the target, i.e. exit point
At the point in time T0, we calculate the signal and get S0. At the point in time T1 we calculate the signal at the point T0 and obtain S1. No false signals is when for any T0 and T1 at T1 > T0, S0 = S1.
- This is actually the default.
What exactly are you suggesting? You could write in a private message...
General conclusion: it makes no sense to consider entry points without exit points.
Regarding reversal.
Global trends and their reversal depend on fundamental factors. It is unrealistic to trade them with small deposits.
These trends are formed by big players (there are not so many of them).
Using trend entries after corrections is possible.
If we assume that intraday corrections are formed by average players to make profit on small ones, then it is possible to use SOT indicators to separate deals of small ones from other players.
Mass accumulation of orders on levels and there are reversal places after a trend correction.
In the 3rd wave we may catch 2 corrective reversals and we may determine them by the sharp acceleration of price movement with low volumes at the levels, which are described in the classics of tehanalysis.
I don't think I have told you any news.
I want to discuss the oscillator period from this - andreybs 26.06.2011 13:05 - post. - Can I post the same screenshot, where the oscillator period will be half as long?