Indicators

Technical indicators are calculations over a run of candles that traders read as signs of trend, momentum or volatility. This page lists the 52 indicator methods every instrument inherits, in five groups that follow TA-Lib's own grouping. Each method fetches the instrument's candles through prices, runs one TA-Lib function over them, and returns the candles with one or more columns added, or None when UBI has no candles for the range.

The table below shows the five groups and what each is for.

Group	Class	Methods	What it measures
Overlap studies	OverlapStudies	12	Averages and bands drawn on the same scale as the price
Momentum indicators	MomentumIndicators	28	How fast and how strongly the price is moving
Volume indicators	VolumeIndicators	3	Whether volume confirms the price's moves
Volatility indicators	VolatilityIndicators	3	How far the price ranges from candle to candle
Cycle indicators	CycleIndicators	6	Repeating cycles, found with the Hilbert transform

The flowchart below is a rough guide to which group answers which question, with one commonly used method from each.

flowchart LR
    Q{"What do you<br/>want to know?"}
    Q -->|"Which way is the trend?"| O["Overlap studies<br/>simple_moving_average"]
    Q -->|"Is the move running out of steam?"| M["Momentum indicators<br/>relative_strength_index"]
    Q -->|"Is volume behind the move?"| V["Volume indicators<br/>on_balance_volume"]
    Q -->|"How far does it swing?"| A["Volatility indicators<br/>average_true_range"]
    Q -->|"Is it trending or cycling?"| C["Cycle indicators<br/>hilbert_transform_trend_mode"]

Arguments every method shares

Every method on this page also takes the five range arguments of prices, which are interval, from_date, to_date, days and adjusted. They are left out of the tables below to keep them readable, and Analysis describes them. Pass every argument by keyword, as in simple_moving_average(window=20, days=365), because the argument order is not the same from method to method.

The tables use the library's argument names, which spell out TA-Lib's. The table below maps the ones that differ most.

This library	TA-Lib
window	timeperiod
fast_period, slow_period, signal_period	fastperiod, slowperiod, signalperiod
moving_average_type	matype
fast_k_period, slow_k_period, slow_d_period, fast_d_period	fastk_period, slowk_period, slowd_period, fastd_period
standard_deviations_up, standard_deviations_down	nbdevup, nbdevdn
volume_factor	vfactor
fast_limit, slow_limit	fastlimit, slowlimit

Several methods take a moving average type, an integer that chooses which kind of average TA-Lib uses inside the calculation. The values are TA-Lib's own, as the table below lists; 0, a simple moving average, is the default everywhere.

Value	Moving average
0	Simple
1	Exponential
2	Weighted
3	Double exponential
4	Triple exponential
5	Triangular
6	Kaufman adaptive
7	MESA adaptive
8	Triple generalised double exponential (T3)

The methods

The sections below hold one table per group, in the same order as the table at the top of the page. The "Columns added" column gives each new column's name, in which a placeholder such as <window> is replaced by the argument's value, so relative_strength_index(window=14) adds rsi_14.

Overlap studies

Overlap studies are drawn on the same scale as the price, so they can be plotted over the candles. Moving averages smooth the price, Bollinger bands put an envelope around it, and the parabolic SAR marks a trailing stop level.

Method	What it adds	Own arguments and defaults	TA-Lib function	Columns added
simple_moving_average	Adds the simple moving average of one candle column	window=10, column='close'	SMA	sma_<window>
exponential_moving_average	Adds the exponential moving average of one candle column	window=10, column='close'	EMA	ema_<window>
bollinger_bands	Adds the upper, middle and lower Bollinger bands of one candle column	window=10, standard_deviations_up=2, standard_deviations_down=2, column='close'	BBANDS	bb_upper_<window>, bb_middle_<window>, bb_lower_<window>
weighted_moving_average	Adds the weighted moving average of one candle column	window=10, column='close'	WMA	wma_<window>
double_exponential_moving_average	Adds the double exponential moving average of one candle column	window=10, column='close'	DEMA	dema_<window>
triple_exponential_moving_average	Adds Tillson's T3 triple exponential moving average of one candle column	window=10, volume_factor=0.7, column='close'	T3	t3_<window>
kaufman_adaptive_moving_average	Adds the Kaufman adaptive moving average of one candle column	window=10, column='close'	KAMA	kama_<window>
mesa_adaptive_moving_average	Adds the MESA adaptive moving average and its following average of one candle column	fast_limit=0.5, slow_limit=0.05, column='close'	MAMA	mama, fama
triangular_moving_average	Adds the triangular moving average of one candle column	window=10, column='close'	TRIMA	trima_<window>
parabolic_sar	Adds the parabolic stop and reverse from the high and low columns	acceleration=0.02, maximum=0.2	SAR	psar
mid_point	Adds the midpoint of the highest and lowest value of one candle column over each window	window=10, column='close'	MIDPOINT	mid_point_<window>
middle_price	Adds the midpoint of the highest high and lowest low over each window	window=10	MIDPRICE	middle_price_<window>

triple_exponential_moving_average calls TA-Lib's T3, Tillson's T3, rather than TEMA, which is the indicator most charting tools call the triple exponential moving average. The name was kept from the old project, and its column is t3_<window>.

Momentum indicators

Momentum indicators measure how fast the price is moving and whether that speed is growing or fading. Most of them are oscillators that move between fixed bounds, such as the relative strength index between 0 and 100.

Method	What it adds	Own arguments and defaults	TA-Lib function	Columns added
moving_average_convergence_divergence	Adds the moving average convergence divergence line, its signal line and their difference	fast_period=12, slow_period=26, signal_period=9, column='close'	MACD	macd_<fast>_<slow>_<signal>, macd_<fast>_<slow>_<signal>_signal, macd_<fast>_<slow>_<signal>_hist
average_directional_movement_index	Adds the average directional movement index	window=14	ADX	adx_<window>
momentum	Adds the momentum of one candle column, its change over each window	window=14, column='close'	MOM	momentum_<window>
commodity_channel_index	Adds the commodity channel index	window=14	CCI	cci_<window>
average_directional_movement_index_rating	Adds the average directional movement index rating	window=10	ADXR	adxr_<window>
absolute_price_oscillator	Adds the absolute price oscillator, the difference between a fast and a slow moving average	fast_period=12, slow_period=26, moving_average_type=0, column='close'	APO	apo_<fast>_<slow>
aroon	Adds the Aroon down and Aroon up lines	window=10	AROON	aroon_down_<window>, aroon_up_<window>
aroon_oscillator	Adds the Aroon oscillator, Aroon up minus Aroon down	window=10	AROONOSC	aroon_osc_<window>
balance_of_power	Adds the balance of power	none	BOP	bop
chande_momentum_oscillator	Adds the Chande momentum oscillator of one candle column	window=10, column='close'	CMO	cmo_<window>
directional_movement_index	Adds the directional movement index	window=10	DX	dx_<window>
moving_average_convergence_divergence_extended	Adds the moving average convergence divergence with a chosen moving average type for each of its three averages	fast_period=12, fast_moving_average_type=0, slow_period=26, slow_moving_average_type=0, signal_period=9, signal_moving_average_type=0, column='close'	MACDEXT	macd_<fast>_<slow>_<signal>, macd_signal_<fast>_<slow>_<signal>, macd_hist_<fast>_<slow>_<signal>
money_flow_index	Adds the money flow index, a relative strength index weighted by volume	window=14	MFI	mfi_<window>
minus_directional_indicator	Adds the minus directional indicator	window=14	MINUS_DI	minus_di_<window>
minus_directional_movement	Adds the minus directional movement	window=14	MINUS_DM	minus_dm_<window>
plus_directional_indicator	Adds the plus directional indicator	window=14	PLUS_DI	plus_di_<window>
plus_directional_movement	Adds the plus directional movement	window=14	PLUS_DM	plus_dm_<window>
percentage_price_oscillator	Adds the percentage price oscillator, the gap between a fast and a slow moving average as a percentage	fast_period=12, slow_period=26, moving_average_type=0, column='close'	PPO	ppo<fast>_<slow>
rate_of_change	Adds the rate of change of one candle column as a percentage	window=14, column='close'	ROC	roc_<window>
rate_of_change_percent	Adds the rate of change of one candle column as a fraction	window=14, column='close'	ROCP	rocp_<window>
rate_of_change_ratio	Adds the rate of change of one candle column as a ratio	window=14, column='close'	ROCR	rocr_<window>
relative_strength_index	Adds the relative strength index of one candle column	window=14, column='close'	RSI	rsi_<window>
stochastic_oscillator	Adds the slow stochastic oscillator's %K and %D lines	fast_k_period=5, slow_k_period=3, slow_k_moving_average_type=0, slow_d_period=3, slow_d_moving_average_type=0	STOCH	slowk_<slow_k_period>, slowd_<slow_d_period>
stochastic_fast_oscillator	Adds the fast stochastic oscillator's %K and %D lines	fast_k_period=5, fast_d_period=3, fast_d_moving_average_type=0	STOCHF	stochf_fastk<fast_k_period>, stochf_fastd<fast_d_period>
stochastic_relative_strength_index	Adds the stochastic relative strength index's %K and %D lines for one candle column	window=14, fast_k_period=5, fast_d_period=3, fast_d_moving_average_type=0, column='close'	STOCHRSI	stochrsi_fastk<fast_k_period>, stochrsi_fastd<fast_d_period>
trix	Adds TRIX, the rate of change of a triple smoothed exponential moving average of one candle column	window=15, column='close'	TRIX	trix_<window>
ultimate_oscillator	Adds the ultimate oscillator, which blends buying pressure over three windows	fast_period=7, slow_period=14, signal_period=28	ULTOSC	ultosc_<fast>_<slow>_<signal>
williams_percent_r	Adds Williams %R	window=14	WILLR	willr_<window>

A few labels are irregular, and they are kept as the old project wrote them so that existing code reading the frames still works. The two MACD methods name their signal and histogram columns differently, ppo, stochf_ and stochrsi_ labels have no underscore before their numbers, and ultimate_oscillator calls its short, middle and long windows fast_period, slow_period and signal_period. The directional movement defaults also differ: average_directional_movement_index uses 14 candles, while average_directional_movement_index_rating and directional_movement_index use 10.

stochastic_relative_strength_index takes a separate window for the length of the underlying RSI, 14 by default, and passes fast_k_period to TA-Lib's fastk_period. With its defaults it matches talib.STOCHRSI with TA-Lib's own defaults.

Volume indicators

Volume indicators combine price and traded volume, to show whether buying or selling pressure is behind a move. They need a volume column, so they are only meaningful for instruments whose candles carry real volume.

Method	What it adds	Own arguments and defaults	TA-Lib function	Columns added
chaikin_accumulation_distribution_line	Adds the Chaikin accumulation distribution line	none	AD	chaikin_ad
chaikin_accumulation_distribution_oscillator	Adds the Chaikin accumulation distribution oscillator	fast_period=3, slow_period=10	ADOSC	chaikin_adosc<fast>_<slow>
on_balance_volume	Adds the on balance volume, measured against one candle column	column='close'	OBV	obv

Volatility indicators

Volatility indicators measure how far the price ranges. The true range of a candle is the largest of its high minus its low and the gaps from the previous close, and the average true range smooths it over a window.

Method	What it adds	Own arguments and defaults	TA-Lib function	Columns added
average_true_range	Adds the average true range	window=14	ATR	atr_<window>
normalized_average_true_range	Adds the average true range as a percentage of the close	window=14	NATR	natr<window>
true_range	Adds each candle's true range	none	TRANGE	tr

The normalised average true range's column is natr<window>, without the underscore that atr_<window> has.

Cycle indicators

Cycle indicators use John Ehlers' Hilbert transform to find a repeating cycle in the price and say whether the market is trending or cycling. They take no window, because the transform chooses its own.

Method	What it adds	Own arguments and defaults	TA-Lib function	Columns added
hilbert_transform_dominant_cycle_period	Adds the Hilbert transform dominant cycle period of one candle column	column='close'	HT_DCPERIOD	ht_dcperiod
hilbert_transform_dominant_cycle_phase	Adds the Hilbert transform dominant cycle phase of one candle column	column='close'	HT_DCPHASE	ht_dcphase
hilbert_transform_phasor_components	Adds the Hilbert transform in-phase and quadrature phasor components of one candle column	column='close'	HT_PHASOR	inphase, quadrature
hilbert_transform_sine_wave	Adds the Hilbert transform sine wave and lead sine wave of one candle column	column='close'	HT_SINE	sine, lead_sine
hilbert_transform_trend_mode	Adds the Hilbert transform trend mode of one candle column, 1 in a trend and 0 in a cycle	column='close'	HT_TRENDMODE	ht_trendmode
hilbert_transform_trend_line	Adds the Hilbert transform instantaneous trend line of one candle column	column='close'	HT_TRENDLINE	ht_trendline

These need a long run of candles before their first value. TA-Lib 0.6.8 reports a lookback of 32 candles for HT_DCPERIOD and HT_PHASOR, and 63 for the other four, so a short range gives columns that are mostly or entirely empty.

A worked example

The example below computes a 14-day relative strength index for the nearest MCX gold future. It was run when the commodities module was checked on 2026-09-20, against a local UBI, and returned 61 rows. The output shows the three rows that check kept, with only the datetime, close and rsi_14 columns.

PythonOutput

from tradingmachine.assets import commodities

expiries = commodities.CommodityFutures.expiries(exchange="mcx", underlying_symbol="GOLD")
gold = commodities.CommodityFutures(exchange="mcx", underlying_symbol="GOLD", expiry_date=expiries[0])

frame = gold.relative_strength_index(window=14, days=90)
print(frame[["datetime", "close", "rsi_14"]])

                 datetime    close    rsi_14
2026-09-14 00:00:00+05:30 151230.0 44.259329
2026-09-15 00:00:00+05:30 150809.0 43.351621
2026-09-16 00:00:00+05:30 152470.0 47.892248

An RSI below 30 is conventionally read as oversold and one above 70 as overbought, so gold's readings in the mid-forties that week were neutral.

Because every method returns the candles, you can combine two indicators by joining their new columns. The next example, which was not run for this page, keeps the days on which a share closed below its lower Bollinger band while its RSI was under 30. Each method fetches the candles again, but both requests ask for the same range, so the rows line up.

from tradingmachine.assets import equities

reliance = equities.Equity(exchange="nse", symbol="RELIANCE")

bands = reliance.bollinger_bands(window=20, days=365)
strength = reliance.relative_strength_index(window=14, days=365)

frame = bands.merge(strength[["datetime", "rsi_14"]], on="datetime")
oversold = frame[(frame["close"] < frame["bb_lower_20"]) & (frame["rsi_14"] < 30)]
print(oversold[["datetime", "close", "bb_lower_20", "rsi_14"]])

Remember that either call can return None, for an instrument or range with no candles, so real code checks for that before merging.