Volatility API Reference

Volatility Measurement Functions

This module provides functions for measuring volatility in financial time series data.

atr(close, high, low, period=14)

Calculate Average True Range (ATR) over a specified period.

ATR measures market volatility by decomposing the entire range of an asset price.

Parameters:

Name	Type	Description	Default
close	ndarray	Close prices	required
high	ndarray	High prices. If None, assumes close contains close prices and high=low=close	required
low	ndarray	Low prices. If None, assumes close contains close prices and high=low=close	required
period	int	Number of periods to calculate ATR	14

Returns:

Type	Description
ndarray	ATR values

bollinger_bands(data, period=20, std_dev=2.0)

Calculate Bollinger Bands over a specified period.

Bollinger Bands consist of a middle band (SMA), an upper band (SMA + kstd), and a lower band (SMA - kstd).

Parameters:

Name	Type	Description	Default
data	ndarray	Input time series data	required
period	int	Number of periods for the moving average	20
std_dev	float	Number of standard deviations for the upper and lower bands	2.0

Returns:

Type	Description
tuple of numpy.ndarray	Tuple containing (upper_band, middle_band, lower_band)

donchian_channels(data, high, low, period=20)

Calculate Donchian Channels over a specified period.

Donchian Channels consist of an upper band (highest high), a lower band (lowest low), and a middle band (average of upper and lower).

Parameters:

Name	Type	Description	Default
data	ndarray	Input time series data (typically close prices)	required
high	ndarray	High prices. If None, uses data	required
low	ndarray	Low prices. If None, uses data	required
period	int	Number of periods for the channels	20

Returns:

Type	Description
tuple of numpy.ndarray	Tuple containing (upper_channel, middle_channel, lower_channel)

historical_volatility(data, period=21, annualization_factor=252)

Calculate historical volatility over a specified period.

Historical volatility is the standard deviation of log returns, typically annualized.

Parameters:

Name	Type	Description	Default
data	ndarray	Input time series data	required
period	int	Number of periods to calculate volatility	21
annualization_factor	int	Factor to annualize volatility (252 for daily data, 52 for weekly, 12 for monthly)	252

Returns:

Type	Description
ndarray	Historical volatility values as percentage

keltner_channels(close, high, low, period=20, atr_period=10, multiplier=2.0)

Calculate Keltner Channels over a specified period.

Keltner Channels consist of a middle band (EMA), an upper band (EMA + kATR), and a lower band (EMA - kATR).

Parameters:

Name	Type	Description	Default
close	ndarray	Close prices	required
high	ndarray	High prices. If None, assumes close contains close prices and high=low=close	required
low	ndarray	Low prices. If None, assumes close contains close prices and high=low=close	required
period	int	Number of periods for the EMA	20
atr_period	int	Number of periods for the ATR	10
multiplier	float	Multiplier for the ATR	2.0

Returns:

Type	Description
tuple of numpy.ndarray	Tuple containing (upper_channel, middle_channel, lower_channel)

volatility_ratio(data, period=21, smooth_period=5)

Calculate Volatility Ratio over a specified period.

Volatility Ratio compares recent volatility to historical volatility. Values above 1 indicate increasing volatility, values below 1 indicate decreasing volatility.

Parameters:

Name	Type	Description	Default
data	ndarray	Input time series data	required
period	int	Number of periods for historical volatility	21
smooth_period	int	Number of periods to smooth the ratio	5

Returns:

Type	Description
ndarray	Volatility Ratio values