QuanTAlib/docs/indicators.md

Indicator Catalog

QuanTAlib provides technical indicators organized into mathematical families. Understanding these families helps choose the right tool for the analytical problem at hand. Selecting an indicator without understanding its category leads to confusion at best, losses at worst.

Category Reference

Category	What It Measures	Representative Indicators	When to Reach for It
Trends (FIR)	Trend direction via finite impulse response filters	SMA, WMA, ALMA, HMA, LSMA	Trend identification with predictable lag and finite memory. Output depends only on a fixed window of past prices.
Trends (IIR)	Trend direction via infinite impulse response filters	EMA, DEMA, TEMA, JMA, KAMA, MAMA	Trend identification with recursive calculation and theoretically infinite memory. More responsive per unit of smoothness.
Filters	Signal processing filters for noise reduction	Bessel, Butterworth, Super Smoother	Removing noise while preserving trend structure. Designed by engineers, borrowed by traders.
Oscillators	Cyclical movement around a baseline	RSI, MACD, AO, UltOsc	Identifying overbought/oversold conditions and potential reversals. Bounded indicators that oscillate.
Dynamics	Trend strength and structural changes	ADX, Aroon, SuperTrend, Chop	Determining market regime (trending vs ranging) and measuring trend conviction.
Momentum	Speed and magnitude of price changes	Momentum, ROC, Velocity	Measuring acceleration or deceleration in price. First derivative territory.
Volatility	Size and variability of price movements	ATR, StdDev, Bollinger Bands	Position sizing, stop-loss placement, regime identification. How much prices move matters as much as direction.
Volume	Trading activity and price-volume relationships	OBV, VWAP, A/D	Confirming price movements with participation. Volume validates or contradicts price action.
Channels	Price boundaries and range definitions	Donchian, Keltner, Bollinger	Breakout strategies and range-bound trading. Defining "normal" so abnormal becomes visible.
Statistics	Mathematical relationships between price series	Correlation, Covariance, Beta, Z-Score	Portfolio analysis, pairs trading, statistical arbitrage. Quantitative analysis beyond single instruments.
Numerics	Mathematical transformations and signal processing	Convolution, Integration, Differentiation	Custom indicator development and advanced signal processing. Building blocks for novel indicators.
Errors	Measurement accuracy and model fit quality	MAE, RMSE, Residuals, R²	Model validation and forecast assessment. Quantifying wrongness before production quantifies losses.
Forecasts	Future price prediction and projection	Linear regression extrapolation, adaptive prediction	Projecting price based on historical patterns. Predictions that invite humility.
Cycles	Periodic patterns and dominant frequencies	Hilbert Transform, Dominant Cycle	Identifying cyclical market behavior. Markets exhibit cycles; detecting them reliably remains hard.

Selection by Experience Level

Beginning technical analysis? Start with Trends (FIR/IIR), Volatility, and Oscillators. SMA teaches moving average fundamentals. ATR teaches volatility measurement. RSI teaches bounded oscillators. These provide foundation for everything else.

Building a trading strategy? Add Volume for confirmation, Channels for breakouts, Dynamics for regime detection. Volume validates price moves. Channels define breakout boundaries. ADX distinguishes trending from ranging markets.

Quantitative development? Filters, Statistics, Numerics, and Errors provide tools for signal processing, model validation, and custom indicator construction. Butterworth filters for noise reduction. Correlation for pairs trading. Error metrics for model selection.

Implemented Indicators

Trends (FIR)

Finite Impulse Response filters. Output depends only on a fixed window of inputs. Always stable. Predictable lag characteristics.

Indicator	Full Name	Notes
ALMA	Arnaud Legoux MA	Gaussian-weighted with offset parameter
BLMA	Blackman Window MA	Spectral leakage reduction
BWMA	Bessel-Weighted MA	Linear phase response
CONV	Convolution MA	Arbitrary kernel support
DWMA	Double Weighted MA	WMA applied twice
GWMA	Gaussian Weighted MA	Normal distribution weights
HAMMA	Hamming Weighted MA	Spectral analysis window
HANMA	Hanning Weighted MA	Cosine-based window
HMA	Hull MA	Reduced lag via WMA differencing
HWMA	Holt-Winters MA	Triple exponential smoothing
LSMA	Least Squares MA	Linear regression endpoint
PWMA	Pascal Weighted MA	Binomial coefficient weights
SGMA	Savitzky-Golay MA	Polynomial smoothing
SINEMA	Sine-Weighted MA	Sinusoidal weight distribution
SMA	Simple MA	Equal weights, the baseline
TRIMA	Triangular MA	Double-smoothed SMA
WMA	Weighted MA	Linear weight decay

Trends (IIR)

Infinite Impulse Response filters. Output depends on current input and past outputs. Recursive structure. More responsive but requires stability analysis.

Indicator	Full Name	Notes
DEMA	Double Exponential MA	EMA of EMA with lag compensation
DSMA	Deviation-Scaled MA	Volatility-adaptive smoothing
EMA	Exponential MA	The fundamental IIR filter
FRAMA	Fractal Adaptive MA	Dimension-based adaptation
HEMA	Hull Exponential MA	Hull concept with EMA
HTIT	Hilbert Instantaneous Trend	Dominant cycle extraction
JMA	Jurik MA	Adaptive, low-lag, proprietary algorithm
KAMA	Kaufman Adaptive MA	Efficiency ratio adaptation
MAMA	MESA Adaptive MA	Homodyne discriminator based
MMA	Modified MA	Smoothed EMA variant
MGDI	McGinley Dynamic	Market-speed tracking
QEMA	Quad Exponential MA	Four-stage exponential
RGMA	Recursive Gaussian MA	Gaussian approximation
REMA	Regularized Exponential MA	Regularization for stability
RMA	WildeR MA	Wilder's smoothing (1/n decay)
T3	Tillson T3 MA	Six-stage DEMA variant
TEMA	Triple Exponential MA	Three-stage lag reduction
VAMA	Volatility Adjusted MA	ATR-based adaptation
VIDYA	Variable Index Dynamic	CMO-based adaptation
YZVAMA	Yang-Zhang Vol Adjusted	YZ volatility adaptation
ZLEMA	Zero-Lag Exponential MA	Momentum-compensated EMA

Filters

Signal processing filters adapted for financial time series. Designed to separate signal from noise with controlled frequency response.

Indicator	Full Name	Notes
BESSEL	Bessel Filter	Maximally flat group delay
BILATERAL	Bilateral Filter	Edge-preserving smoothing
BPF	BandPass Filter	Frequency band isolation
BUTTER	Butterworth Filter	Maximally flat passband
CHEBY1	Chebyshev Type I	Steeper rolloff with ripple
SSF	Super Smooth Filter	Ehlers two-pole design
USF	Ultimate Smoother	Ehlers high-fidelity filter

Oscillators

Bounded indicators that oscillate around a centerline or between fixed extremes. Useful for mean-reversion signals.

Indicator	Full Name	Notes
AO	Awesome Oscillator	Midpoint momentum
APO	Absolute Price Oscillator	EMA difference
MACD	MACD	EMA crossover system
RSI	Relative Strength Index	Bounded 0-100 momentum
ULTOSC	Ultimate Oscillator	Multi-timeframe weighted

Dynamics

Indicators measuring trend strength, regime, and directional movement quality.

Indicator	Full Name	Notes
ADX	Average Directional Index	Trend strength 0-100
ADXR	ADX Rating	Smoothed ADX
AMAT	Archer MA Trends	MA-based trend detection
AROON	Aroon	High/low recency
AROONOSC	Aroon Oscillator	Aroon Up minus Down
DMX	Jurik DMX	Enhanced directional movement
SUPER	SuperTrend	ATR-based trend bands

Momentum

Rate of change and velocity measurements. First derivatives of price.

Indicator	Full Name	Notes
BOP	Balance of Power	Close position in range
CFB	Composite Fractal Behavior	Jurik fractal momentum
ROC	Rate of Change	Absolute price change over N periods
RSX	Jurik RSX	Smoothed RSI variant
VEL	Jurik Velocity	Adaptive velocity

Volatility

Measures of price variability and range. Essential for position sizing and stop placement.

Indicator	Full Name	Notes
ADR	Average Daily Range	Simple range averaging
ATR	Average True Range	Gap-adjusted range
ATRP	ATR Percent	Normalized ATR

Volume

Price-volume relationships and accumulation/distribution measurements.

Indicator	Full Name	Notes
ADL	Accumulation/Distribution	Volume-weighted close position
ADOSC	Chaikin A/D Oscillator	ADL momentum

Channels

Price envelope and boundary indicators for breakout and mean-reversion strategies.

Indicator	Full Name	Notes
ABBER	Aberration Bands	Statistical deviation bands
ACCBANDS	Acceleration Bands	Volatility-adjusted envelope
DCHANNEL	Donchian Channels	Highest-high / lowest-low breakout bands
DECAYCHANNEL	Decay Min-Max Channel	Exponential decay toward midpoint
FCB	Fractal Chaos Bands	Williams fractal-based support/resistance
JBANDS	Jurik Adaptive Envelope Bands	Snap-to-extreme, decay-to-price volatility bands
KCHANNEL	Keltner Channel	EMA with ATR bands; smoother than Bollinger
MAENV	Moving Average Envelope	Fixed percentage bands around selectable MA type
MMCHANNEL	Min-Max Channel	Rolling highest high / lowest low; O(1) monotonic deques
PCHANNEL	Price Channel	Highest high / lowest low; identical to Donchian
REGCHANNEL	Linear Regression Channel	Linear regression line with standard deviation bands
SDCHANNEL	Standard Deviation Channel	Moving average with standard deviation bands

Statistics

Mathematical and statistical computations on price series.

Indicator	Full Name	Notes
CMA	Cumulative Moving Average	Expanding window average
COVARIANCE	Covariance	Joint variability
LINREG	Linear Regression	Best-fit line
MEDIAN	Rolling Median	50th percentile
SKEW	Skewness	Distribution asymmetry
STDDEV	Standard Deviation	Dispersion measure
SUM	Rolling Sum	Windowed sum
VARIANCE	Variance	Squared deviation

Forecasts

Predictive indicators and extrapolation methods.

Indicator	Full Name	Notes
AFIRMA	Adaptive FIR MA	Predictive FIR filter

Numerics

Mathematical transformations and derivative indicators. Building blocks for analysis.

Indicator	Full Name	Notes
ACCEL	Acceleration (2nd Derivative)	Change in slope; momentum
CHANGE	Percentage Change	Relative price movement (current - past) / past
EXPTRANS	Exponential Transform	e^x transform for log-space reversal
HIGHEST	Rolling Maximum	O(1) amortized via monotonic deque
JERK	Jerk (3rd Derivative)	Change in acceleration
LINEARTRANS	Linear Transform	y = ax + b scaling transformation
LOGTRANS	Logarithmic Transform	Natural log for percentage analysis
LOWEST	Rolling Minimum	O(1) amortized via monotonic deque
MIDPOINT	Rolling Midpoint	(Highest + Lowest) / 2
NORMALIZE	Min-Max Normalization	Scale to [0,1] via rolling min/max
RELU	Rectified Linear Unit	max(0, x); activation function
SIGMOID	Logistic Function	1/(1+e^-x); bounded [0,1] transform
SLOPE	Slope (1st Derivative)	Rate of change; velocity
SQRTTRANS	Square Root Transform	√x; variance to standard deviation conversion

Errors

Error metrics and loss functions for model evaluation, forecast assessment, and strategy validation. Quantifying wrongness before production quantifies losses.

Indicator	Full Name	Notes
WRMSE	Weighted Root Mean Squared Error	Custom observation weighting for error emphasis