OPEN-SOURCE SCRIPT

Isotonic Regression [LuxAlgo]

The Isotonic Regression indicator provides a monotonic fit to price data, ensuring the resulting line is either non-decreasing or non-increasing over a specified lookback period. This tool is particularly useful for identifying underlying trends and significant price plateaus without the lagging or "overshooting" common in standard moving averages or linear regressions.

Note: This indicator calculates its values based on a historical lookback window, which means the regression line will repaint as new bars are added and the window shifts.

🔶 USAGE

Isotonic regression is used to find the best-fitting line to a set of data points under the constraint that the line must move in a specific direction (always up or always down). This creates a "staircase" effect where the model alternates between trending segments and flat plateaus.

🔹 Direction Modes

The indicator offers three ways to determine the fit direction:

Auto: Automatically detects the trend by comparing the start and end prices of the lookback period. If the end price is higher, it fits a non-decreasing line; otherwise, it fits a non-increasing line.
Non-Decreasing: Forces the fit to only move upwards or stay flat, ideal for analyzing bullish structures.
Non-Increasing: Forces the fit to only move downwards or stay flat, ideal for analyzing bearish structures.

🔹 Flat Period Extensions

One of the most powerful features of isotonic regression is the identification of "blocks" or price levels where the trend pauses. When the duration of such a plateau exceeds the Flat Period Threshold, the indicator extends a dashed horizontal line to the current bar, highlighting potential support or resistance levels derived from the regression model.

🔶 DETAILS

The script implements the Pool Adjacent Violators Algorithm (PAVA), which is the standard method for computing isotonic regression.

The algorithm works by partitioning the data into "blocks." If a subsequent data point violates the monotonicity constraint (e.g., price drops in a "non-decreasing" fit), the algorithm pools the current block with the previous one and calculates a weighted average. This process repeats until the entire sequence is monotonic.

For performance efficiency, the indicator utilizes the

Pine Script®

function to render the regression line as a single continuous object rather than hundreds of individual line segments.

🔶 SETTINGS

Lookback Length: The number of bars used to calculate the isotonic fit.
Source: The price data used for the calculation (default is Close).
Direction: Sets the monotonicity constraint (Auto, Non-Decreasing, or Non-Increasing).
Flat Period Threshold: The minimum number of bars a price plateau must last to be highlighted with an extension line.

🔹 Style

Fit Bullish/Bearish: Colors for the regression line based on the detected trend.
Fit Style/Width: Controls the visual representation (Solid, Dashed, Dotted) and thickness of the main regression line.
Ext Bullish/Bearish: Colors for the flat period extension levels.
Ext Style/Width: Controls the visual representation and thickness of the plateau extensions.

오픈 소스 스크립트

트레이딩뷰의 진정한 정신에 따라, 이 스크립트의 작성자는 이를 오픈소스로 공개하여 트레이더들이 기능을 검토하고 검증할 수 있도록 했습니다. 작성자에게 찬사를 보냅니다! 이 코드는 무료로 사용할 수 있지만, 코드를 재게시하는 경우 하우스 룰이 적용된다는 점을 기억하세요.

LuxAlgo

All content & scripts provided by LuxAlgo are for educational purposes only and are not financial advice.

면책사항

해당 정보와 게시물은 금융, 투자, 트레이딩 또는 기타 유형의 조언이나 권장 사항으로 간주되지 않으며, 트레이딩뷰에서 제공하거나 보증하는 것이 아닙니다. 자세한 내용은 이용 약관을 참조하세요.

오픈 소스 스크립트

LuxAlgo

All content & scripts provided by LuxAlgo are for educational purposes only and are not financial advice.