Donchian Channels

I. What is the Donchian Channels Indicator

Donchian Channels is a technical analysis indicator invented by American futures trading pioneer Richard Donchian (1905—1993). Richard Donchian is known as the “Father of Trend Following” and was one of the earliest practitioners to systematically study trend-following trading, profoundly influencing the development of the entire futures industry and quantitative trading field.

Donchian Channels belongs to the Volatility / Bands overlay indicator category and is plotted directly on the price chart. It consists of three lines:

• Upper Band: The highest high over the past $n$ periods
• Lower Band: The lowest low over the past $n$ periods
• Middle Band: The average of the upper and lower bands

The default period is $n = 20$ (corresponding to approximately one month of trading days).

---

II. Mathematical Principles and Calculation

2.1 Core Formulas

Let the high price series be $H_1, H_2, \ldots, H_t$, the low price series be $L_1, L_2, \ldots, L_t$, and the lookback period be $n$.

Upper Band:

$$\text{Upper}_t = \max(H_{t-n+1}, H_{t-n+2}, \ldots, H_t)$$

Lower Band:

$$\text{Lower}_t = \min(L_{t-n+1}, L_{t-n+2}, \ldots, L_t)$$

Middle Band:

$$\text{Middle}_t = \frac{\text{Upper}_t + \text{Lower}_t}{2}$$

2.2 Channel Width

Channel Width can be used to measure volatility:

$$\text{Width}_t = \text{Upper}_t - \text{Lower}_t$$

Normalized channel width:

$$\text{Width\%}_t = \frac{\text{Upper}_t - \text{Lower}_t}{\text{Middle}_t} \times 100$$

2.3 Calculation Steps

1. Determine the lookback period $n$ (default 20)
2. Take the maximum of the high prices over the most recent $n$ trading days -> upper band
3. Take the minimum of the low prices over the most recent $n$ trading days -> lower band
4. Average the upper and lower bands -> middle band

2.4 Fundamental Differences from Bollinger Bands

Feature	Donchian Channels	Bollinger Bands
Upper band calculation	Highest high in the range	SMA + K x Standard Deviation
Lower band calculation	Lowest low in the range	SMA - K x Standard Deviation
Volatility measurement	Price extreme range	Statistical standard deviation
Sensitivity to extreme values	Very high (a single extreme value determines the boundary)	Moderate (extreme values are diluted through averaging)
Channel shape	Staircase-like (changes only when new highs/lows occur)	Smooth curve

---

III. Python Implementation

import numpy as np
import pandas as pd

def donchian_channels(high: pd.Series, low: pd.Series,
                      period: int = 20) -> pd.DataFrame:
    """
    Calculate Donchian Channels.

    Parameters:
        high   : high price series
        low    : low price series
        period : lookback period, default 20

    Returns:
        DataFrame with columns: upper, lower, middle, width
    """
    # Upper band: maximum of n-period highs
    upper = high.rolling(window=period).max()

    # Lower band: minimum of n-period lows
    lower = low.rolling(window=period).min()

    # Middle band: average of upper and lower
    middle = (upper + lower) / 2.0

    # Channel width
    width = upper - lower

    return pd.DataFrame({
        'upper': upper,
        'lower': lower,
        'middle': middle,
        'width': width
    })


def donchian_signals(df: pd.DataFrame, period: int = 20,
                     exit_period: int = 10) -> pd.DataFrame:
    """
    Generate Turtle Trading entry/exit signals based on Donchian Channels.

    Parameters:
        df          : DataFrame with high, low, close columns
        period      : entry channel period (default 20)
        exit_period : exit channel period (default 10)

    Returns:
        DataFrame with channel values and trading signals
    """
    # Entry channel (exclude current bar; use previous day's channel values)
    entry_upper = df['high'].rolling(window=period).max().shift(1)
    entry_lower = df['low'].rolling(window=period).min().shift(1)

    # Exit channel
    exit_upper = df['high'].rolling(window=exit_period).max().shift(1)
    exit_lower = df['low'].rolling(window=exit_period).min().shift(1)

    # Signal generation
    signals = pd.Series('Hold', index=df.index)
    signals[df['close'] > entry_upper] = 'Breakout Long'
    signals[df['close'] < entry_lower] = 'Breakout Short'
    signals[(df['close'] < exit_upper) & (signals.shift(1) == 'Breakout Long')] = 'Long Exit'
    signals[(df['close'] > exit_lower) & (signals.shift(1) == 'Breakout Short')] = 'Short Exit'

    return pd.DataFrame({
        'close': df['close'],
        'entry_upper': entry_upper,
        'entry_lower': entry_lower,
        'exit_upper': exit_upper,
        'exit_lower': exit_lower,
        'signal': signals
    })


# ============ Usage Example ============
if __name__ == '__main__':
    np.random.seed(42)
    n_days = 100

    # Generate simulated OHLCV data
    base_price = 100 + np.cumsum(np.random.randn(n_days) * 0.8)
    df = pd.DataFrame({
        'open':   base_price + np.random.randn(n_days) * 0.3,
        'high':   base_price + np.abs(np.random.randn(n_days) * 1.2),
        'low':    base_price - np.abs(np.random.randn(n_days) * 1.2),
        'close':  base_price,
        'volume': np.random.randint(1000, 10000, n_days)
    })

    # Calculate Donchian Channels
    dc = donchian_channels(df['high'], df['low'], period=20)
    result = pd.concat([df[['close']], dc], axis=1)

    print("=== Donchian Channels Results (last 10 rows) ===")
    print(result.tail(10).to_string())

    # Turtle Trading signals
    sig = donchian_signals(df, period=20, exit_period=10)
    print("\n=== Trading Signals (last 15 rows) ===")
    print(sig.tail(15).to_string())

    print("\n=== Signal Statistics ===")
    print(sig['signal'].value_counts())

---

IV. Problems the Indicator Solves

4.1 Trend Breakout Identification

The most fundamental application of Donchian Channels is breakout trading:

• Price breaks above the upper band: Sets an $n$-day new high, generating a long signal
• Price breaks below the lower band: Sets an $n$-day new low, generating a short signal

This is the purest trend-following logic — a new high means bullish forces dominate, and a new low means bearish forces dominate.

4.2 Turtle Trading System

The classic Turtle Trading system uses dual Donchian Channels:

System	Entry	Exit
System 1	20-day breakout entry	10-day breakout exit
System 2	55-day breakout entry	20-day breakout exit

4.3 Support and Resistance Identification

• The upper band naturally forms a dynamic resistance level ($n$-day high)
• The lower band naturally forms a dynamic support level ($n$-day low)
• The middle band can serve as a reference baseline for trend direction

4.4 Volatility Assessment

Channel width reflects the price fluctuation range over the most recent $n$ days:

• Channel narrows -> Market enters consolidation, volatility decreases
• Channel widens -> Market becomes active, volatility increases

---

V. Advantages, Disadvantages, and Use Cases

Advantages

Advantage	Description
Extremely simple	Requires only high prices, low prices, and one period parameter — zero calculation complexity
Unambiguous	A breakout is a breakout — no gray area
Captures major trends	New highs and new lows are the most fundamental definition of a trend, capturing all significant moves
Extensively validated	The Turtle Trading experiment proved its long-term effectiveness
No repainting	Once a signal is generated, it never disappears

Disadvantages

Disadvantage	Description
Poor performance in ranging markets	Frequently generates false breakout signals in range-bound markets
Staircase-like changes	Upper and lower bands are not smooth and may jump suddenly
Sensitive to single extreme values	A single large bullish or bearish candle can change the channel boundary
Delayed entry	Must wait for a new high/low to enter, potentially missing the early part of a trend
Slow exit signals	Using reverse breakouts for exits can give back substantial profits

Use Cases

• Best suited for: Strongly trending markets (commodity futures, forex trend moves)
• Moderately suited for: Medium to long-term trend-following systems
• Not suited for: Short-term intraday trading; highly choppy markets

Parameter Selection Guidelines

Trading Style	Entry Period	Exit Period
Short-term	10	5
Medium-term (Turtle System 1)	20	10
Long-term (Turtle System 2)	55	20