[GYTS] Ultimate Smoother (3-poles + 2 poles)Ultimate Smoother (3-pole)
🌸 Part of GoemonYae Trading System (GYTS) 🌸
🌸 --------- INTRODUCTION --------- 🌸
💮 Release of 3-Pole Ultimate Smoother
This indicator presents a new 3-pole version of John Ehlers' Ultimate Smoother (2024) . This results in an unconventional filter that exhibits effectively zero lag in practical trading applications, regardless of the set period. By using a 2-pole high-pass filter in its design, it responds to price direction changes on the same bar, while still allowing the user to control smoothness.
💮 What is the Ultimate Smoother?
The original Ultimate Smoother is a revolutionary filter designed by John Ehlers (2024) that smooths price data with virtually zero lag in the pass band. While conventional filters always introduce lag when removing market noise, the Ultimate Smoother maintains phase alignment at low frequencies while still providing excellent noise reduction.
💮 Mathematical Foundation
The Ultimate Smoother achieves its remarkable properties through a clever mathematical approach:
1. Instead of directly designing a low-pass filter (like traditional moving averages), it subtracts a high-pass filter from an all-pass filter (the original input data).
2. At very low frequencies, the high-pass filter contributes almost nothing, so the output closely matches the input in both amplitude and phase.
3. At higher frequencies, the high-pass filter's response increasingly matches the input data, resulting in cancellation through subtraction.
The 3-pole version extends this principle by using a higher-order high-pass filter, requiring additional coefficients and handling more terms in the numerator of the transfer function.
🌸 --------- USAGE GUIDE --------- 🌸
💮 Period Parameter Behaviour
The period parameter in the 3-pole Ultimate Smoother works somewhat counterintuitively:
- Longer periods: Result in less smooth, but more responsive following of the price. The filter output more closely tracks the input data.
- Shorter periods: Produce smoother output but may exhibit overshooting (extrapolating price movement) for larger movements.
This is different from most filters where longer periods typically produce smoother outputs with more lag.
💮 When to Choose 3-Pole vs. 2-Pole
- Choose the 3-pole version when you need zero-lag but want to control the smoothness
- Choose the 2-pole version when you are okay with some lag with the benefit of more smoothness.
🌸 --------- ACKNOWLEDGEMENTS --------- 🌸
This indicator builds upon the pioneering work of John Ehlers, particularly from his article April 2024 edition of TASC's Traders' Tips . The original version is published on TradingView by @PineCodersTASC .
This 3-pole extension was developed by @GoemonYae . Feedback is highly appreciated!
Goemonyaetradingsystem
ColourUtilitiesLibrary "ColourUtilities"
Utility functions for colour manipulation
adjust_colour(rgb, desaturation_amount, transparency_amount)
to reduce saturation or increase transparency of an RGB colour
Parameters:
rgb (color)
desaturation_amount (float) : 0 means no desaturation (colours remains as-is), and 1 means full desaturation (colour turns grey). Can also be used inversely with negative numbers
transparency_amount (float) : How much more transparent the default transparency should become. E.g. with a value of 0.5, a transparency of 0 becomes 50 and 40 becomes 70. A value of 1 makes it fully transparent, en -1 fully opaque.
Returns: color with adjusted saturation and transparency
method apply_default_palette(self, palette_name)
Some nice looking colour palettes, consisting of 6 gradient colours, are already defined here and can be quickly applied to the Palette class
Namespace types: Palette
Parameters:
self (Palette)
palette_name (string) : Currently there are 4 6-coloured palettes available: "GYTS flux signal", "GYTS purple", "GYTS flux filter" and "GYTS maroon"
Returns: None, as it populates the Palette class with pre-defined colours
method get_colour(self, colour_no, transparency)
Retrieves colour from the palette and possibly changes transparency if set
Namespace types: Palette
Parameters:
self (Palette)
colour_no (int) : from the palette
transparency (int) : to possibly change the default transparency of the palette
Returns: colour
method get_dynamic_colour(self, x, mid_point, colour_lb, colour_ub, trend_lookback, use_rate)
Retrieves a colour based on strength and direction of the passed series
Namespace types: Palette
Parameters:
self (Palette)
x (float) : the input data series
mid_point (float) : value as a cutoff point where the bullish/bearish colour scenario
colour_lb (float) : value (lower bound) where to apply the bearish colour at full strength
colour_ub (float) : value (upper bound) where to apply the bullish colour at full strength
trend_lookback (int) : how much bars back to check if there was a consistent move into a certain direction, otherwise a the neutral colour from the centre of the palette will be used.
use_rate (bool) : whether to use the rate (proportional difference with previous `x` value) or the input series `x` directly
Returns: colour
Palette
Fields:
transparency (series__integer)
palette (array__color)
