skfolio.descriptor.DaysToCover#

class skfolio.descriptor.DaysToCover(half_life=21.0, min_periods=None)[source]#

Exponentially weighted days-to-cover descriptor.

Computes the ratio of shares sold short to exponentially weighted average daily volume:

\[ \begin{aligned} \text{EWMA\_volume}(t) &= \lambda \cdot \text{EWMA\_volume}(t-1) + (1 - \lambda) \cdot \text{adj\_volume}(t) \\[0.75em] \text{days\_to\_cover}(t) &= \frac{\text{short\_interest}(t)}{\text{EWMA\_volume}(t)} \end{aligned} \]

where \(\lambda = \exp(-\ln(2) / \text{half\_life})\) is the EWMA decay factor.

Days to cover measures how many trading days it would take short sellers to buy back their positions at the current trading rate. High values indicate crowded short positions relative to liquidity [1] [2].

EWMA smoothing is preferred over a fixed rolling average because daily volume can spike around earnings, index rebalances or news events. EWMA dampens these spikes gradually, producing more stable factor exposures.

Parameters:

half_lifefloat, default=21.0

EWMA half-life in observations for volume smoothing. With daily data, common choices are:

half_life=21: ~1 month
half_life=63: ~3 months
half_life=252: ~1 year

min_periodsint, optional

Minimum number of valid positive-volume observations required for each asset. Until an asset reaches this count, its output is NaN. This warm-up period avoids exposing early EWMA values before the volume estimate has sufficiently converged from its zero initialization. If None, defaults to \(\lceil\text{half\_life}\rceil\), with a minimum of 1.

Attributes:

n_assets_int: Number of assets seen during fitting.
asset_names_ndarray of shape (n_assets,): Asset names seen during fitting.
days_to_cover_ndarray of shape (n_assets,): Last days-to-cover value for each asset.

Methods

`fit_transform`(X[, y])	Compute exponentially weighted days to cover.
`get_metadata_routing`()	Get metadata routing of this object.
`get_params`([deep])	Get parameters for this estimator.
`partial_fit_transform`(X[, y])	Update state and return days to cover for this batch.
`set_params`(**params)	Set the parameters of this estimator.

See also

ShortInterest: Short interest as fraction of shares outstanding.
EWShareTurnover: EWMA share turnover (volume / shares outstanding).

Notes

short_interest is the number of shares held short. Non-missing values must be finite and non-negative.

adj_volume is split-adjusted trading volume. Non-missing values must be finite and non-negative.

The EWMA state is updated only for positive-volume observations. NaN or zero adj_volume holds the EWMA state and does not increment the valid-observation count. NaN short_interest propagates to the output but does not prevent the volume state from updating.

The active_mask property of the AssetPanel distinguishes holidays from delistings.

References

[1]

“An investigation of the informational role of short interest in the Nasdaq market” The Journal of Finance. Desai, H., Ramesh, K., Thiagarajan, S. R., & Balachandran, B. V. (2002).

[2]

“Short interest, institutional ownership, and stock returns” Journal of Financial Economics. Asquith, P., Pathak, P. A., & Ritter, J. R. (2005).

Examples

>>> from skfolio.datasets import make_synthetic_characteristics
>>> from skfolio.descriptor import DaysToCover
>>>
>>> X = make_synthetic_characteristics()
>>>
>>> # 1-month EWMA volume smoothing (default)
>>> descriptor = DaysToCover()
>>> days_to_cover = descriptor.fit_transform(X)
>>>
>>> # 3-month EWMA volume smoothing
>>> descriptor = DaysToCover(half_life=63)
>>> days_to_cover = descriptor.fit_transform(X)

fit_transform(X, y=None, **fit_params)[source]#

Compute exponentially weighted days to cover.

Parameters:

XAssetPanel: Input panel containing short_interest and adj_volume.
yNone: Ignored. Present for compatibility with scikit-learn’s API.
**fit_paramsdict: Additional fit parameters. Ignored.

Returns:

days_to_coverndarray of shape (n_observations, n_assets): Short interest divided by EWMA-smoothed daily volume for each observation and asset.

get_metadata_routing()#

Get metadata routing of this object.

Please check User Guide on how the routing mechanism works.

Returns:

routingMetadataRequest: A MetadataRequest encapsulating routing information.

get_params(deep=True)#

Get parameters for this estimator.

Parameters:

deepbool, default=True: If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns:

paramsdict: Parameter names mapped to their values.

partial_fit_transform(X, y=None, **fit_params)[source]#

Update state and return days to cover for this batch.

This method supports online updates by continuing from the current fitted state. Use fit_transform to start from a clean state.

Parameters:

XAssetPanel: Input panel containing short_interest and adj_volume.
yNone: Ignored. Present for compatibility with scikit-learn’s API.
**fit_paramsdict: Additional fit parameters. Ignored.

Returns:

days_to_coverndarray of shape (n_observations, n_assets): Short interest divided by EWMA-smoothed daily volume for each observation and asset.

set_params(**params)#

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters:

**paramsdict: Estimator parameters.

Returns:

selfestimator instance: Estimator instance.