Category: Turning Points
Abbreviation: PeakprobB
Input parameters
Name
|
Setting
|
Default
|
High
|
High time series
|
High
|
Low
|
Low time series
|
Low
|
Close
|
Close time series
|
Close
|
Neighborhood
|
Integer >= 1
|
3
|
Window
|
Integer >= 1
|
50
|
Calculations
The Peak Probability (bars) indicator output is the probability that the current bar is at the next turning point peak, based on the valley-to-peak bar distance mean and standard deviation in the Window. For definitions of turning point peak and turning point valley, see help topic Overview.
The probability ranges from 0 (peak is not probable) to 1 (peak is extremely probable). The peak probability only calculates if the most recent turning point in the window is a valley. If it is a peak, then the peak probability is set to zero. For statistically reliable probabilities the window must be large enough compared to the neighborhood parameter. Generally, the ratio Window:Neighborhood of at least 20:1or higher is highly desirable.
This indicator can be used in a trading strategy as a Long Exit (or Short Entry) signal. Make a trading rule in the form of a relational operator A>B, where A=PeakprobB and B is a threshold, ranging from 0 to 1. Let the Trader’s optimizer find the best threshold B, the neighborhood size, and the window size that maximize your trading strategy return.
|