QuanTAlib/lib/momentum/cci/Cci.md

CCI - Commodity Channel Index

Property	Value
Category	Momentum
Inputs	OHLCV bar (TBar)
Parameters	period (default 20)
Outputs	Single series (CCI)
Output range	Varies (see docs)
Warmup	period bars

TL;DR

• The Commodity Channel Index (CCI) is a versatile momentum-based oscillator developed by Donald Lambert in 1980.
• Parameterized by period (default 20).
• Output range: Varies (see docs).
• Requires period bars (20 default) of warmup before first valid output (IsHot = true).
• Validated against TA-Lib, Skender, and Tulip reference implementations where available.

Overview

The Commodity Channel Index (CCI) is a versatile momentum-based oscillator developed by Donald Lambert in 1980. Originally designed for commodity trading, it measures the deviation of price from its statistical mean, normalized by mean absolute deviation.

Formula

TP = (High + Low + Close) / 3
SMA = Simple Moving Average of TP over period
Mean Deviation = SUM(|TP - SMA|) / period
CCI = (TP - SMA) / (0.015 × Mean Deviation)

Key Characteristics

Property	Value
Default Period	20
Lambert Constant	0.015
Returns	Unbounded oscillator (typically -300 to +300)
Warmup Period	Equal to period
Input Data	OHLC bars (uses Typical Price)

Signal Interpretation

Primary Levels

• Above +100: Strong uptrend, potentially overbought
• Below -100: Strong downtrend, potentially oversold
• Zero Line: Centerline crossover indicates trend change

Trading Strategies

1. Trend Identification: Values consistently above/below zero indicate trend direction
2. Overbought/Oversold: Extreme readings (+200/-200) suggest reversal potential
3. Divergence: Price making new high/low while CCI fails to confirm
4. Zero-Line Cross: Bullish when crossing above, bearish when crossing below

Lambert Constant (0.015)

The 0.015 constant was chosen by Lambert to ensure that approximately 70-80% of CCI values fall between +100 and -100 under normal market conditions. This provides a statistical framework where:

• Values outside ±100 indicate significant price movement
• Extended readings suggest strong trends
• Extreme values (±200 or beyond) are relatively rare

Usage

Basic Construction

// Create CCI with default 20-period
var cci = new Cci();

// Create CCI with custom period
var cci = new Cci(14);

Streaming Updates

foreach (var bar in realTimeData)
{
    TValue result = cci.Update(bar);
    double cciValue = result.Value;
    
    if (cciValue > 100)
        Console.WriteLine("Overbought territory");
    else if (cciValue < -100)
        Console.WriteLine("Oversold territory");
}

Batch Processing

TSeries results = Cci.Batch(barSeries, period: 20);

Comparison with Other Oscillators

Indicator	Bounds	Best For
CCI	Unbounded	Trend strength, divergence
RSI	0-100	Overbought/oversold levels
Stochastic	0-100	Price position within range

Historical Context

• Developed by Donald Lambert (1980)
• Originally published in Commodities magazine
• Early application for identifying cyclical trends in commodities
• Now widely used across all asset classes

References

• Lambert, D.R. (1980). "Commodity Channel Index: Tool for Trading Cyclic Trends"
• TradingView CCI Documentation
• Investopedia CCI Guide

Performance Profile

Operation Count (Streaming Mode)

Each Update() call on CCI(N) performs a full O(N) mean-deviation scan over the ring buffer. There is no closed-form running-sum decomposition for mean absolute deviation — the absolute values prevent the cancellation that makes SMA or variance incremental. The RingBuffer manages the sliding window; computing MAD requires visiting every element.

Operation	Count	Cost (cycles)	Subtotal
Ring buffer push	1	3	~3
Running sum update (TP add/subtract)	2	1	~2
SMA divide	1	8	~8
MAD scan: N subtractions + N ABS	2N	2	~2N·2
MAD divide	1	8	~8
Final scale + divide (0.015×MAD)	2	3	~6
Total	2N + 7	—	~(4N + 27) cycles

O(N) streaming cost per bar. For the default N = 20: ~107 cycles. No incremental shortcut exists for MAD; SIMD vectorization of the scan loop is the primary optimization lever.

Batch Mode (SIMD Analysis)

Operation	Vectorizable?	Notes
Typical price (H+L+C)/3	Yes	3-wide FMA, AVX2 vectorizable
Rolling SMA via prefix sums	Yes	VADDPD on register array
MAD inner loop (ABS + accumulate)	Yes	VABSPD + VADDPD, width-8 per AVX2 lane
Final CCI scale	Yes	scalar multiply after reduction
State dependency across bars (SMA)	Partial	prefix sum removes dependency; MAD is fully independent per bar

AVX2 processes 4 doubles per instruction. For the inner MAD loop of N=20, that is 5 SIMD passes vs 20 scalar iterations — roughly 3× throughput gain. The outer bar loop remains SIMD-friendly since each bar's TP is independent once the window positions are known.