Commodity Channel Index (CCI)

I. What is CCI

The Commodity Channel Index (CCI) is a momentum oscillator that measures how far the price has deviated from its statistical average. It calculates the deviation between the Typical Price and its moving average, normalized by the Mean Deviation, to identify the beginning and end of trends as well as overbought/oversold conditions.

Historical Background

CCI was published by Donald Lambert in 1980 in Commodities magazine. Despite the “Commodity” in its name, CCI is applicable to all financial markets including stocks, forex, and cryptocurrencies. Lambert designed CCI with the original purpose of identifying cyclical turning points in commodity prices.

Indicator Classification

• Type: Oscillator, displayed in a separate panel
• Category: Momentum
• Default Parameters: Period $n = 20$
• Value Range: Unbounded, typically oscillates between -200 and +200

---

II. Mathematical Principles and Calculation

Core Formulas

Step 1: Calculate Typical Price

$TP_t = \frac{H_t + L_t + C_t}{3}$

Step 2: Calculate SMA of Typical Price

$SMA(TP)_t = \frac{1}{n}\sum_{i=0}^{n-1} TP_{t-i}$

Step 3: Calculate Mean Deviation

$MD_t = \frac{1}{n}\sum_{i=0}^{n-1} |TP_{t-i} - SMA(TP)_t|$

Step 4: Calculate CCI

$CCI_t = \frac{TP_t - SMA(TP)_t}{0.015 \times MD_t}$

Step-by-Step Calculation Logic

1. Calculate Typical Price: (High + Low + Close) / 3 for each bar
2. Calculate SMA(20) of TP: Simple average of the past 20 TP values
3. Calculate Mean Deviation: Average of the absolute differences between each TP and the SMA over the past 20 periods
4. Normalize: (Current TP - SMA) / (0.015 x MD)

Value Interpretation

CCI Value	Meaning
> +100	Price significantly above statistical average, potential overbought or strong uptrend
0	Price exactly at the statistical average
< -100	Price significantly below statistical average, potential oversold or strong downtrend

---

III. Python Implementation

import numpy as np
import pandas as pd


def cci(high: pd.Series,
        low: pd.Series,
        close: pd.Series,
        period: int = 20) -> pd.Series:
    """
    Calculate the Commodity Channel Index (CCI)

    Parameters
    ----------
    high : pd.Series
        High price series
    low : pd.Series
        Low price series
    close : pd.Series
        Close price series
    period : int
        Calculation period, default 20

    Returns
    -------
    pd.Series
        CCI values
    """
    tp = (high + low + close) / 3

    tp_sma = tp.rolling(window=period, min_periods=period).mean()

    # Calculate mean deviation
    md = tp.rolling(window=period, min_periods=period).apply(
        lambda x: np.mean(np.abs(x - x.mean())), raw=True
    )

    cci_values = (tp - tp_sma) / (0.015 * md)

    return cci_values


# ========== Usage Example ==========
if __name__ == "__main__":
    np.random.seed(42)
    dates = pd.date_range("2024-01-01", periods=120, freq="D")
    base = 100 + np.cumsum(np.random.randn(120) * 1.0)

    df = pd.DataFrame({
        "date": dates,
        "open":  base + np.random.randn(120) * 0.3,
        "high":  base + np.abs(np.random.randn(120) * 0.8),
        "low":   base - np.abs(np.random.randn(120) * 0.8),
        "close": base,
        "volume": np.random.randint(1000, 10000, size=120),
    })
    df.set_index("date", inplace=True)

    # Calculate CCI
    df["CCI_20"] = cci(df["high"], df["low"], df["close"], period=20)

    print("=== CCI Results (last 15 rows) ===")
    print(df[["close", "CCI_20"]].tail(15).to_string())

    # Signal generation
    df["signal"] = np.where(
        df["CCI_20"] > 100, 1,       # Strong zone -> Trend long
        np.where(df["CCI_20"] < -100, -1, 0)  # Weak zone -> Trend short
    )

    # Crossover of +100/-100
    df["zone"] = np.sign(df["CCI_20"].clip(-100, 100).replace(0, np.nan)).ffill()
    cross_up = (df["CCI_20"] > 100) & (df["CCI_20"].shift(1) <= 100)
    cross_down = (df["CCI_20"] < -100) & (df["CCI_20"].shift(1) >= -100)

    print("\n=== CCI Crosses Above +100 ===")
    print(df.loc[cross_up, ["close", "CCI_20"]].to_string())
    print("\n=== CCI Crosses Below -100 ===")
    print(df.loc[cross_down, ["close", "CCI_20"]].to_string())

---

IV. Problems the Indicator Solves

1. Trend Identification and Confirmation

CCI can be used to identify the start of new trends:

• CCI crosses above +100 — Price breaks above the statistically normal range, potentially starting a new uptrend
• CCI crosses below -100 — Price breaks below the statistically normal range, potentially starting a new downtrend

2. Overbought / Oversold

Traditional oscillator usage:

Condition	Signal
CCI drops back below +100 from above	Sell (overbought reversion)
CCI rises back above -100 from below	Buy (oversold reversion)

3. Divergence Signals

• Bearish Divergence: Price new high + CCI fails to make new high — upward momentum weakening
• Bullish Divergence: Price new low + CCI fails to make new low — downward momentum weakening

4. Zero Line Analysis

• CCI operating above zero — price above statistical average, bullish bias
• CCI operating below zero — price below statistical average, bearish bias

---

V. Advantages, Disadvantages, and Use Cases

Advantages

Advantage	Description
Unbounded	Unlike RSI’s 0-100 limit, CCI can reflect extreme deviations
Solid statistical foundation	Normalization via mean absolute deviation is statistically sound
Dual purpose	Can be used for both trend following and mean reversion
Uses HLC	Combines high, low, and close price information

Disadvantages

Disadvantage	Description
Ambiguous signals	Unbounded range makes extreme value definition less clear than RSI
Mode selection required	Choosing between trend-following vs. mean-reversion adds subjectivity
Unstable in extremes	When MD is very small, CCI values can be extremely amplified
Single period	Default 20-period may not suit all markets

Use Cases

• Commodity futures: CCI was originally designed for commodities and excels with cyclical instruments
• Trend breakouts: CCI’s first breach of +/-100 can signal trend initiation
• Cyclical markets: Most effective on assets with clear cyclical patterns

Comparison with Related Indicators

Comparison	CCI	RSI	Stochastic
Value Range	Unbounded	0-100	0-100
Normalization	Mean Deviation	Wilder Smoothing	Price Range
Usage Mode	Trend + Mean Reversion	Mainly Mean Reversion	Mainly Mean Reversion
Data Input	H/L/C	C	H/L/C