quantpylib.hft.features

flow_pnl(trades)

Calculates the pnl of the a set of trades.

Parameters:

Name	Type	Description	Default
trades	ndarray	2d-array of trades as [[timestamp, price, size, direction], ...].	required

Returns:

Type	Description
	np.ndarray: The timestamps, running PnLs and cumulative volumes.

rolling_vol(mids, n, exp=False)

Calculate rolling volatility of the mid prices.

Parameters:

Name	Type	Description	Default
mids	ndarray	Array of mid prices.	required
n	int	The size of the rolling window.	required
exp	bool	Whether to use exponential weighting. Defaults to False.	False

Returns:

Type	Description
	np.ndarray: An array of rolling volatilities.

trade_imbalance(trades, decay_function=lambda sample: exponential_weights(arr=sample, unique_values=True, normalize=True))

Calculate the trade imbalance for a given set of trades w.r.t. some decay function.

Parameters:

Name	Type	Description	Default
trades	ndarray	2d-array of trades as [[timestamp, price, size, direction], ...].	required
decay_function	function	The decay function. Defaults to exponential_weights.	lambda sample: exponential_weights(arr=sample, unique_values=True, normalize=True)

vol(sample, exp=False)

Calculate the sample volatility.

Parameters:

Name	Type	Description	Default
sample	ndarray	The input sample.	required
exp	bool	Whether to use exponential weighting. Defaults to False.	False

Returns:

Name	Type	Description
float		The calculated sample volatility.