balipour

Impossible RSI [BAPig]

The Impossible RSI is built on original RSI . If users want traditional RSI value, they can choose "Traditional RSI" in RSI mode. Untick "Log transform" ,"Error Correcting Smoothing " and "show columns" options in settings.

Orginal RSI allows negative price as input. By using the log transform on input price, we eliminate the negative price scenario.
The reverse RSI demonstrates the negative price scenario.

Error-correcting smoothing is a proprietary smoothing filter, it uses signal processing techniques to create a low lag, smooth, low overshoot smoothing method. User can change the smoothness by adjusting the length in "Low pass length" setting.

By Selecting "Gaussian RSI" option in "RSI Mode", the user can have a close to Gaussian distribution oscillator. We use fisher transform to transform the distribution of traditional rsi into gaussian. It has sharper turning points. We recommend users that use mean reversion strategy to use "Gaussian RSI". Most Oscillators doesn't have a gaussian distribution , therefore the overbought and oversold level doesn't have statistical significance.
Distribution of Gaussian RSI

By Selecting "Zero Mean RSI" option in "RSI Mode", the user can have an oscillator with a consistent zero mean. A high pass filter is applied to filter out the long cycle and trend component. It makes the oscillator zero mean and reduces lag. Oscillators usually stuck in overbought or oversold when the trend is too strong. High pass filter gets rid of this situation. The period in high pass filter represents the longest trend period we are interested in. We recommend users that use trend-following strategy, for example, buy or sell based on oscillator cross the zero line to use "Zero Mean RSI".

Traditional RSI and Gaussian RSI doesn't have a zero mean. It has skewness in the distribution, therefore the zero-crossing strategy is not useful. (Most Oscillators has that problem, the mean is not centred at zero).

Mean of "Zero Mean RSI"


By Selecting "Zero Mean Gaussian RSI", the user gets a zero mean gaussian distributed oscillator. We use both high pass filter and fisher transformation here. This is a perfect combination for an oscillator. This RSI mode is most recommended.
Distribution of "Zero Mean Gaussian RSI"

Users can adjust the length of MA to have a trigger line

Users can adjust the length of the band to have a dynamic overbought and oversold level, a longer length is recommended. When "Gaussian RSI " or "Zero mean Gaussian RSI" mode is selected, the bands are statistically significant. The default setting of standard deviation is 1.96 which is 95% confidence interval.

Short Length Band

When "Show Columns" option is tick off. The oscillator will have a 0 to 100 range like the traditional RSI value. The mean is no longer zero. However, it only changes the scale. It doesn't change the shape or slope.

We used an error-correcting smoothing technique, fisher transform , log transformation and high pass filter on traditional RSI to make it a smooth, low lag, zero mean, near gaussian distributed oscillator. It's an ideal oscillator almost impossible to beat.





Versionshinweise: bug fixed

Our Website🅱🅰🐷: https://bapigindicator.com

Sign up on our website to get access to our indicators.

Join Our Discord for further discussions:

https://discord.gg/ink
Skript nur auf Einladung

Der Zugriff auf dieses Skript ist auf vom Autor autorisierte User beschränkt und normalerweise kostenpflichtig. Sie können es zu Ihren Favoriten hinzufügen, aber Sie können es nur verwenden, nachdem Sie die Erlaubnis angefordert und vom Autor erhalten haben. Kontaktieren Sie balipour für weitere Informationen oder folgen Sie den Anweisungen des Autors unten.

TradingView rät davon ab, für ein Skript zu bezahlen und es zu verwenden, bis Sie dem Autor zu 100% vertrauen und verstehen wie das Skript funktioniert. In vielen Fällen können Sie eine gute Open-Source-Alternative kostenlos in unserer öffentlichen Bibliothek finden.

Hinweise des Autors

bapigindicator.com

Möchten Sie dieses Skript auf einem Chart verwenden?

Warnung: Bitte lesen Sie dies, bevor Sie Zugriff anfordern.