OPEN-SOURCE SCRIPT
업데이트됨 Coppock Curve Multi-Filter [MarkitTick]
💡 A dual-momentum oscillator built on the classic Coppock Curve, extended with an optional adaptive source pre-filter, an ADX strength gate, and a full ATR-based trade-management layer with staged take-profits, on-chart price levels, and a live dashboard. The core wave is a weighted moving average of two rate-of-change readings, but everything measured downstream of that wave — signal timing, trend bias, and risk levels — can be reshaped by up to eight independent, toggleable filters, giving traders a single oscillator that can behave anywhere from "classic long-term Coppock" to a tightly gated, multi-condition entry engine.
✨ Originality and Utility
The stock Coppock Curve is a single-purpose, long-only momentum tool: sum two rate-of-change readings, smooth with a weighted moving average, and watch for crosses above zero. This script keeps that foundation intact but restructures it into a bidirectional signal engine with a stack of independent confirmation layers that the original concept never included.
The key structural change is the adaptive source stage. Rather than feeding raw closing price directly into the rate-of-change calculations, the script offers a choice of eight different smoothing methods — including a custom Kalman Filter estimator and a custom LLAMA (Linear-Lag Adaptive Moving Average) function — that first condition the price series before Coppock's ROC math is applied. This means the character of the entire curve can be tuned from responsive to heavily smoothed without altering the underlying two-ROC-plus-WMA structure that defines the Coppock method.
Layered on top of that are seven optional gating and confirmation mechanisms (ADX strength, divergence, slope acceleration, volume, higher-timeframe alignment, volatility-adjusted zero line, and signal persistence) that traders can combine in any subset. Because each filter operates independently and can be switched on or off, the same core wave can be configured for a slow trend-confirmation approach or a fast, tightly-filtered signal generator, giving the tool a much broader utility range than a standard Coppock plot.
Beyond signal generation, the script converts each qualifying cross into a full trade plan: an ATR-derived stop-loss, three R-multiple take-profit tiers, live price levels drawn on the chart, and a real-time dashboard summarizing bias, filter states, and trade levels — none of which exist in the original Coppock Curve concept or in standard TradingView implementations of it.
🔬 Methodology and Concepts
● Core Wave Construction
The engine begins with an adaptive source stage. If no adaptive filter is selected, the raw chosen source (default: close) feeds directly into the calculation. If a filter is selected, the source is pre-smoothed using one of the following:
Once the (optionally smoothed) source is established, two Rate of Change values are calculated against it — a long lookback and a short lookback, independently configurable. These two ROC values are summed and passed through a weighted moving average, producing the final Coppock Curve value. This is structurally identical to the classic Coppock formula, but with the adaptive pre-filter as an optional intermediate step.
• ADX Strength Filter
When enabled, the script calculates the Directional Movement Index (+DI, -DI, ADX) over a configurable length. A signal — whether a slope change, a cross, or a zero-line cross — is only considered valid if the ADX reading is at or above the user-defined threshold. This filters out Coppock movements that occur during weak or directionless conditions.
• Slope and Cross Detection
The script tracks whether the curve is rising or falling bar-to-bar, and separately detects two types of crosses: a cross of the curve against its own prior value (used as the primary bull/bear signal) and a cross of the curve against the zero line (used as a secondary trend-state signal). Both cross types respect the ADX filter when it is active.
• Signal Locking
A "Lock Signal" input freezes the active signal and trade levels on the most recent bar, preventing new signals from overwriting the currently displayed trade plan — useful for holding a specific setup visible while monitoring live price action.
● Trade-Level Automation
Every new bullish or bearish cross (confirmed and unlocked) triggers a full trade-plan calculation:
These levels persist on the chart until a new opposing signal fires (or, if Lock Signal is active, until manually released), and are dynamically extended to the current bar so the trade plan remains visible in real time. Take-profit and stop labels update their text once price actually touches each respective level, marking it as hit along with the realized percentage move.
● Optional Confirmation Filters
Seven additional filters exist as inputs in the script but should be understood as configuration flags a trader can layer onto the core signal logic depending on their own methodology:
Traders should treat these as intended-purpose toggles per their input tooltips and confirm behavior against the ADX filter and core cross logic, which are the two filters fully wired into the signal path in this build.
🎨 Visual Guide
● Main Panel (Separate Pane)
● Price Chart Overlay
● Dashboard Table
A compact table (position configurable) displays, in real time: the current symbol and timeframe, the Lock Signal state, the raw Coppock value, the current bias (Bullish / Bearish / Neutral, color-coded), the individual long and short ROC readings, whether the curve is currently above or below zero, and — when trade levels are enabled — the live Entry, SL, TP1, TP2, and TP3 prices. If the ADX filter is active, its current reading is shown alongside a pass/fail color cue. If an adaptive filter is selected, its name is displayed for quick reference.
📖 How to Use
⚙️ Inputs and Settings
• Core Settings
• Filters
• Trade Tools
• Visuals
• Dashboard
• Alerts
• Colors
🔍 Deconstruction of the Underlying Scientific and Academic Framework
● Rate of Change and the Coppock Curve
The foundation of this script is Edwin Coppock's original curve, published in Barron's in 1962, which sums a long-term and a short-term Rate of Change and smooths the result with a weighted moving average. Rate of Change itself is a first-order momentum measure — the percentage difference between the current value and its value N bars ago — rooted in the broader technical-analysis principle that the velocity of price change often leads price direction itself. Coppock's original design used a WMA specifically because it weights recent data more heavily than a simple average while remaining less reactive to single-bar noise than an exponential average.
● Weighted and Hull Moving Averages
The Weighted Moving Average used both in the final smoothing stage and optionally in the adaptive pre-filter assigns linearly decreasing weights to older data points, a technique long used to balance responsiveness against noise rejection. The Hull Moving Average, developed by Alan Hull, extends this idea by combining WMAs of different lengths in a way designed to reduce lag while preserving smoothness — a documented refinement of the general weighted-average family.
● Kalman Filtering
The Kalman Filter, originally developed by Rudolf Kálmán in the context of control and estimation theory, is a recursive algorithm for estimating an unknown value from a series of noisy observations. In this implementation, the filter maintains a running estimate and an error term, computes a Kalman gain from the ratio of predicted error to total error each bar, and updates the estimate by blending new price data in proportion to that gain. This gives the estimate more responsiveness when its own uncertainty is high and more smoothness when uncertainty is low — the same estimation principle underlying Kalman's original work, applied here to a single noisy input series rather than a multi-variable state system.
● Directional Movement and Trend Strength (Wilder)
The optional ADX filter is built on J. Welles Wilder's Directional Movement System, which derives +DI and -DI from directional price movement smoothed with Wilder's own moving average technique, then compresses their divergence into the Average Directional Index (ADX) as a bounded measure of trend strength independent of direction. Using ADX as a gating condition reflects the broader academic distinction between trend-following and mean-reverting market regimes — Wilder's system was explicitly designed to help separate the two.
● Average True Range and Volatility-Based Risk Sizing
Stop-loss and take-profit distances in this script are derived from Average True Range, also introduced by Wilder, which measures volatility by accounting for gaps as well as intraperiod range. Sizing risk as a multiple of ATR — rather than a fixed point or percentage value — is a widely documented approach in position-sizing literature because it scales stop distance to the instrument's actual recent volatility rather than an arbitrary constant.
● R-Multiples and Risk-Reward Structuring
The three-tiered take-profit structure expresses reward as a multiple of initial risk (an "R-multiple"), a framework popularized in trading risk-management literature to normalize outcomes across trades of different sizes and volatility regimes, allowing performance to be evaluated in terms of risk-adjusted return rather than raw price movement.
⚠️ Disclaimer
All provided scripts and indicators are strictly for educational exploration and must not be interpreted as financial advice or a recommendation to execute trades. We expressly disclaim all liability for any financial losses or damages that may result, directly or indirectly, from the reliance on or application of these tools. Market participation carries inherent risk where past performance never guarantees future returns, leaving all investment decisions and due diligence solely at your own discretion.
✨ Originality and Utility
The stock Coppock Curve is a single-purpose, long-only momentum tool: sum two rate-of-change readings, smooth with a weighted moving average, and watch for crosses above zero. This script keeps that foundation intact but restructures it into a bidirectional signal engine with a stack of independent confirmation layers that the original concept never included.
The key structural change is the adaptive source stage. Rather than feeding raw closing price directly into the rate-of-change calculations, the script offers a choice of eight different smoothing methods — including a custom Kalman Filter estimator and a custom LLAMA (Linear-Lag Adaptive Moving Average) function — that first condition the price series before Coppock's ROC math is applied. This means the character of the entire curve can be tuned from responsive to heavily smoothed without altering the underlying two-ROC-plus-WMA structure that defines the Coppock method.
Layered on top of that are seven optional gating and confirmation mechanisms (ADX strength, divergence, slope acceleration, volume, higher-timeframe alignment, volatility-adjusted zero line, and signal persistence) that traders can combine in any subset. Because each filter operates independently and can be switched on or off, the same core wave can be configured for a slow trend-confirmation approach or a fast, tightly-filtered signal generator, giving the tool a much broader utility range than a standard Coppock plot.
Beyond signal generation, the script converts each qualifying cross into a full trade plan: an ATR-derived stop-loss, three R-multiple take-profit tiers, live price levels drawn on the chart, and a real-time dashboard summarizing bias, filter states, and trade levels — none of which exist in the original Coppock Curve concept or in standard TradingView implementations of it.
🔬 Methodology and Concepts
● Core Wave Construction
The engine begins with an adaptive source stage. If no adaptive filter is selected, the raw chosen source (default: close) feeds directly into the calculation. If a filter is selected, the source is pre-smoothed using one of the following:
- Simple, Exponential, or RMA-based moving averages
- A Double WMA (a weighted moving average applied twice in succession, producing extra lag reduction)
- A Triple VWMA (three successive volume-weighted moving average passes)
- A Hull Moving Average
- A custom LLAMA function, which computes a simple moving average over the lookback window, then adds a linear slope term (calculated from the change in price across the window divided by the window length) scaled by half the window length — effectively projecting the average forward along its own recent trajectory
- A custom Kalman Filter estimator, which maintains a running estimate and error variance, calculates a Kalman gain each bar from the ratio of predicted error to total error, and blends the new price into the estimate proportionally to that gain — placing more weight on new data when the filter's own uncertainty is high, and more weight on the existing estimate when it is low
Once the (optionally smoothed) source is established, two Rate of Change values are calculated against it — a long lookback and a short lookback, independently configurable. These two ROC values are summed and passed through a weighted moving average, producing the final Coppock Curve value. This is structurally identical to the classic Coppock formula, but with the adaptive pre-filter as an optional intermediate step.
• ADX Strength Filter
When enabled, the script calculates the Directional Movement Index (+DI, -DI, ADX) over a configurable length. A signal — whether a slope change, a cross, or a zero-line cross — is only considered valid if the ADX reading is at or above the user-defined threshold. This filters out Coppock movements that occur during weak or directionless conditions.
• Slope and Cross Detection
The script tracks whether the curve is rising or falling bar-to-bar, and separately detects two types of crosses: a cross of the curve against its own prior value (used as the primary bull/bear signal) and a cross of the curve against the zero line (used as a secondary trend-state signal). Both cross types respect the ADX filter when it is active.
• Signal Locking
A "Lock Signal" input freezes the active signal and trade levels on the most recent bar, preventing new signals from overwriting the currently displayed trade plan — useful for holding a specific setup visible while monitoring live price action.
● Trade-Level Automation
Every new bullish or bearish cross (confirmed and unlocked) triggers a full trade-plan calculation:
- Entry is set to the prior bar's close
- Stop-loss is placed at a configurable multiple of ATR away from entry, in the direction opposing the trade
- Three take-profit levels are calculated as configurable R-multiples of the initial risk distance (the entry-to-stop distance), projected in the trade's favor
- Each level's distance from entry is also expressed as a percentage for quick reference
These levels persist on the chart until a new opposing signal fires (or, if Lock Signal is active, until manually released), and are dynamically extended to the current bar so the trade plan remains visible in real time. Take-profit and stop labels update their text once price actually touches each respective level, marking it as hit along with the realized percentage move.
● Optional Confirmation Filters
Seven additional filters exist as inputs in the script but should be understood as configuration flags a trader can layer onto the core signal logic depending on their own methodology:
- Divergence Filter — intended to suppress cross signals that run counter to a detected price/Coppock divergence
- Slope Acceleration Filter — intended to require the curve's slope itself to be increasing, not merely positive, before validating a signal
- Volume Confirmation Filter — intended to require current volume to exceed its moving average before a signal is accepted
- HTF Alignment Filter — intended to require a higher-timeframe Coppock reading to agree with the signal's direction
- Volatility-Adjusted Zero Line — intended to require zero-line crosses to clear a noise band derived from the indicator's own recent volatility, reducing whipsaw signals near the zero line
- Signal Persistence Filter — intended to require the curve's direction to hold for a minimum number of bars before a signal is treated as valid
Traders should treat these as intended-purpose toggles per their input tooltips and confirm behavior against the ADX filter and core cross logic, which are the two filters fully wired into the signal path in this build.
🎨 Visual Guide
● Main Panel (Separate Pane)
- The primary line plot shows the Coppock Curve itself. It is colored using the Bull Color when the curve is rising and the ADX filter (if active) passes, the Bear Color when falling under the same condition, and the Neutral Color otherwise.
- A histogram of the same Coppock value is plotted in columns beneath the line, using a four-tier color scheme: strong bull shading when the curve is above zero and rising, weak bull shading when above zero but not rising, weak bear shading when below zero but rising, and strong bear shading when below zero and falling.
- A dashed horizontal zero line marks the neutral threshold that separates bullish and bearish curve territory.
- Small triangle markers appear directly on the curve at the exact bar where it crosses zero — an upward triangle in Bull Color for an upward zero-cross, and a downward triangle in Bear Color for a downward zero-cross.
● Price Chart Overlay
- When candle coloring is enabled, the price candles themselves are recolored using the same four-tier histogram coloring described above, turning the price chart into a visual heatmap of underlying Coppock strength and direction.
- When a new signal fires and trade levels are enabled, five horizontal lines are drawn directly on price: a solid stop-loss line, a dashed entry line, and three dashed take-profit lines with progressively increasing opacity from TP1 to TP3. Each line carries a right-aligned label showing its role and exact price.
- A shaded "risk zone" fills the area between the stop-loss and entry lines, and a "reward zone" fills the area between the entry and TP3 lines, giving an immediate visual sense of the risk-to-reward geometry of the active trade plan.
- Once a take-profit or stop level is touched by price, its label updates in place to show a hit confirmation along with the realized percentage gain or loss.
● Dashboard Table
A compact table (position configurable) displays, in real time: the current symbol and timeframe, the Lock Signal state, the raw Coppock value, the current bias (Bullish / Bearish / Neutral, color-coded), the individual long and short ROC readings, whether the curve is currently above or below zero, and — when trade levels are enabled — the live Entry, SL, TP1, TP2, and TP3 prices. If the ADX filter is active, its current reading is shown alongside a pass/fail color cue. If an adaptive filter is selected, its name is displayed for quick reference.
📖 How to Use
- Treat a bullish cross (curve turning up) as a potential long-side signal, and a bearish cross (curve turning down) as a potential short-side signal, especially when it aligns with a zero-line cross in the same direction.
- Use the zero line as a broader trend-state filter: readings above zero generally reflect positive intermediate-term momentum, while readings below zero reflect negative momentum, independent of the immediate slope.
- Enable the ADX filter to restrict signals to periods of measurable trend strength, reducing signals generated during flat or choppy conditions.
- Select an adaptive filter method to change the responsiveness of the underlying source feeding the Coppock calculation — faster methods like EMA or the Kalman Filter increase sensitivity, while methods like the Triple VWMA or SMA produce a smoother, slower curve.
- When a signal fires, use the automatically plotted Entry, SL, and TP1–TP3 lines as a starting reference for trade structure, and adjust position sizing according to the displayed stop distance and your own risk tolerances.
- Use candle heatmap coloring as a quick visual scan across the chart to spot where momentum has historically been strongest or weakest, independent of reading the oscillator pane directly.
- Configure the webhook alert action strings in the Alerts group to match the payload keys expected by your automation or webhook receiver before relying on the JSON-formatted alerts for execution.
⚙️ Inputs and Settings
• Core Settings
- Source — the price series the calculation is based on (default: close)
- Long ROC Length — lookback for the long-term rate-of-change component
- Short ROC Length — lookback for the short-term rate-of-change component
- WMA Smoothing Length — window for the final weighted moving average applied to the combined ROC values
• Filters
- Use ADX Filter / ADX Threshold / ADX Length — enables trend-strength gating and configures its sensitivity
- Adaptive Filter / Adaptive Filter Length — selects the pre-smoothing method applied to price before the ROC/WMA math, and its lookback window
- Use Divergence Filter / Divergence Pivot Lookback — configuration for suppressing signals against detected divergence
- Use Slope Acceleration Filter — configuration for requiring accelerating slope before a signal
- Use Volume Confirmation Filter / Volume MA Length — configuration for requiring above-average volume
- Use HTF Alignment Filter / HTF Alignment Timeframe — configuration for requiring higher-timeframe agreement
- Use Volatility-Adjusted Zero Line / Volatility Zero Band Multiple / Volatility Zero Band Length — configuration for a noise-adjusted zero-cross threshold
- Use Signal Persistence Filter / Persistence Bars — configuration for requiring a minimum number of bars of consistent direction
• Trade Tools
- Lock Signal — freezes the currently active signal and trade levels
- SL ATR Multiple — sets stop-loss distance as a multiple of ATR
- TP1 / TP2 / TP3 R-Multiple — sets each take-profit distance as a multiple of the initial risk
- ATR Length — lookback for the Average True Range calculation used in stop placement
- Show Trade Levels — toggles the on-chart lines, labels, and dashboard trade-level rows
• Visuals
- Use Candle Coloring — toggles heatmap-style recoloring of price candles
- Show Histogram — toggles the columned histogram beneath the main curve
- Show Zero-Cross Markers — toggles the triangle markers at zero-line crosses
• Dashboard
- Show Dashboard — toggles the on-chart summary table
- Position — sets the table's screen position
• Alerts
- Action strings for Bull Cross, Bear Cross, Zero Cross Up/Down, Close Long/Short, and TP1/TP2/TP3/SL Hit — these populate the "action" field of each JSON alert payload, allowing the alerts to be mapped directly to webhook or automation logic
• Colors
- Full palette control over bull/bear/neutral coloring, histogram tiers, dashboard styling, and all trade-level line and fill colors
🔍 Deconstruction of the Underlying Scientific and Academic Framework
● Rate of Change and the Coppock Curve
The foundation of this script is Edwin Coppock's original curve, published in Barron's in 1962, which sums a long-term and a short-term Rate of Change and smooths the result with a weighted moving average. Rate of Change itself is a first-order momentum measure — the percentage difference between the current value and its value N bars ago — rooted in the broader technical-analysis principle that the velocity of price change often leads price direction itself. Coppock's original design used a WMA specifically because it weights recent data more heavily than a simple average while remaining less reactive to single-bar noise than an exponential average.
● Weighted and Hull Moving Averages
The Weighted Moving Average used both in the final smoothing stage and optionally in the adaptive pre-filter assigns linearly decreasing weights to older data points, a technique long used to balance responsiveness against noise rejection. The Hull Moving Average, developed by Alan Hull, extends this idea by combining WMAs of different lengths in a way designed to reduce lag while preserving smoothness — a documented refinement of the general weighted-average family.
● Kalman Filtering
The Kalman Filter, originally developed by Rudolf Kálmán in the context of control and estimation theory, is a recursive algorithm for estimating an unknown value from a series of noisy observations. In this implementation, the filter maintains a running estimate and an error term, computes a Kalman gain from the ratio of predicted error to total error each bar, and updates the estimate by blending new price data in proportion to that gain. This gives the estimate more responsiveness when its own uncertainty is high and more smoothness when uncertainty is low — the same estimation principle underlying Kalman's original work, applied here to a single noisy input series rather than a multi-variable state system.
● Directional Movement and Trend Strength (Wilder)
The optional ADX filter is built on J. Welles Wilder's Directional Movement System, which derives +DI and -DI from directional price movement smoothed with Wilder's own moving average technique, then compresses their divergence into the Average Directional Index (ADX) as a bounded measure of trend strength independent of direction. Using ADX as a gating condition reflects the broader academic distinction between trend-following and mean-reverting market regimes — Wilder's system was explicitly designed to help separate the two.
● Average True Range and Volatility-Based Risk Sizing
Stop-loss and take-profit distances in this script are derived from Average True Range, also introduced by Wilder, which measures volatility by accounting for gaps as well as intraperiod range. Sizing risk as a multiple of ATR — rather than a fixed point or percentage value — is a widely documented approach in position-sizing literature because it scales stop distance to the instrument's actual recent volatility rather than an arbitrary constant.
● R-Multiples and Risk-Reward Structuring
The three-tiered take-profit structure expresses reward as a multiple of initial risk (an "R-multiple"), a framework popularized in trading risk-management literature to normalize outcomes across trades of different sizes and volatility regimes, allowing performance to be evaluated in terms of risk-adjusted return rather than raw price movement.
⚠️ Disclaimer
All provided scripts and indicators are strictly for educational exploration and must not be interpreted as financial advice or a recommendation to execute trades. We expressly disclaim all liability for any financial losses or damages that may result, directly or indirectly, from the reliance on or application of these tools. Market participation carries inherent risk where past performance never guarantees future returns, leaving all investment decisions and due diligence solely at your own discretion.
릴리즈 노트
Resolved an issue affecting certain filter modules릴리즈 노트
fix bug오픈 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.
💡 Proprietary indicators. Original research. Built by analysts who trade.
📢 Free Telegram: t.me/MarkitTick_Updates
👑 Premium: markittick.com
📢 Free Telegram: t.me/MarkitTick_Updates
👑 Premium: markittick.com
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
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.
💡 Proprietary indicators. Original research. Built by analysts who trade.
📢 Free Telegram: t.me/MarkitTick_Updates
👑 Premium: markittick.com
📢 Free Telegram: t.me/MarkitTick_Updates
👑 Premium: markittick.com
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.