bayesline.api.equity.FactorModelApi

class bayesline.api.equity.FactorModelApi

Bases: ABC

Methods

__init__()
dates()	Returns All dates covered by this risk model.
estimation_universe(*[, start, end, ...])	Obtains the risk model estimation universe for this risk model.
estimation_universe_as_task(*[, start, end, ...])
exposures(*[, start, end, id_type, stage])	Obtains the risk model exposures for this risk model.
exposures_as_task(*[, start, end, id_type, ...])
factors(*which)	Parameters which: FactorType The factor types to return, e.g. Market, Style, Industry, Region. By default returns all factors.
fret(*[, freq, cumulative, start, end])	Parameters freq: str, optional The frequency of the return aggregation, e.g. D for daily. Defaults to daily (i.e. unaggregated) cumulative: bool, optional If True, returns the cumulative returns. :8: (WARNING/2) Definition list ends without a blank line; unexpected unindent. start: DateLike, optional end: DateLike, optional
fret_as_task(*[, freq, cumulative, start, end])
future_asset_returns(*[, start, end, id_type])	Obtains the asset returns for this risk model on the next day.
future_asset_returns_as_task(*[, start, ...])
market_caps(*[, start, end, id_type])	Obtains the exposure-weighted market caps for this risk model.
market_caps_as_task(*[, start, end, id_type])
p_values()
p_values_as_task()
r2()
r2_as_task()
sigma2()
sigma2_as_task()
t_stats()
t_stats_as_task()
universe(*[, start, end, id_type, stage])	Obtains the risk model universe for this risk model.
universe_as_task(*[, start, end, id_type, stage])
weights(*[, start, end, id_type])	Obtains the idiosynchratic volatility weights for this risk model.
weights_as_task(*[, start, end, id_type])

abstract dates() → list[date]

Returns

All dates covered by this risk model.

abstract factors(*which: Literal['Market', 'Style', 'Industry', 'Region', 'Other']) → list[str]

Parameters

which: FactorType

The factor types to return, e.g. Market, Style, Industry, Region. By default returns all factors.

Returns

list of all factors for the given factor types.

abstract exposures(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None, stage: int = 1) → DataFrame

Obtains the risk model exposures for this risk model.

Parameters

start: DateLike, optional

The start date of the data to return, inclusive.

end: DateLike, optional

The end date of the data to return, inclusive.

id_type: IdType, optional

The id type to return asset ids in, e.g. ticker. The given id type must be supported by the linked universe.

stage: int, default=1,

The stage of the factor model to return exposures for, of the potentially multi-stage regression. Default to 1 (the first stage).

Raises

ValueError

If the given id type is not supported or date range is invalid.

Returns

pl.DataFrame

The data for the given date range with the first two column as the date and asset id. The remaining columns are the individual styles.

abstract exposures_as_task(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None, stage: int = 1) → Task[DataFrame]

abstract universe(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None, stage: int = 1) → DataFrame

Obtains the risk model universe for this risk model.

Parameters

start: DateLike, optional

The start date of the data to return, inclusive.

end: DateLike, optional

The end date of the data to return, inclusive.

id_type: IdType, optional

The id type to return asset ids in, e.g. ticker. The given id type must be supported by the linked universe.

stage: int, default=1,

The stage of the factor model to return universe for, of the potentially multi-stage regression. Default to 1 (the first stage).

Raises

ValueError

If the given id type is not supported or date range is invalid.

Returns

pl.DataFrame

The data for the given date range where the first column is the date and the remaining columns are the asset ids. The values are the universe inclusion.

abstract universe_as_task(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None, stage: int = 1) → Task[DataFrame]

abstract estimation_universe(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None, stage: int = 1) → DataFrame

Obtains the risk model estimation universe for this risk model.

Parameters

start: DateLike, optional

The start date of the data to return, inclusive.

end: DateLike, optional

The end date of the data to return, inclusive.

id_type: IdType, optional

The id type to return asset ids in, e.g. ticker. The given id type must be supported by the linked universe.

stage: int, default=1,

The stage of the factor model to return universe for, of the potentially multi-stage regression. Default to 1 (the first stage).

Raises

ValueError

If the given id type is not supported or date range is invalid.

Returns

pl.DataFrame

The data for the given date range where the first column is the date and the remaining columns are the asset ids. The values are the estimation universe inclusion.

abstract estimation_universe_as_task(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None, stage: int = 1) → Task[DataFrame]

abstract market_caps(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None) → DataFrame

Obtains the exposure-weighted market caps for this risk model.

Parameters

start: DateLike, optional

The start date of the data to return, inclusive.

end: DateLike, optional

The end date of the data to return, inclusive.

id_type: IdType, optional

The id type to return asset ids in, e.g. ticker. The given id type must be supported by the linked universe.

Raises

ValueError

If the given id type is not supported or date range is invalid.

Returns

pl.DataFrame

The data for the given date range where the rows are the total market cap for each date, weighted by the exposure of each asset. For industry factors, this specifically means that the value is the sum of all assets in the estimation universe in that industry.

abstract market_caps_as_task(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None) → Task[DataFrame]

abstract weights(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None) → DataFrame

Obtains the idiosynchratic volatility weights for this risk model.

Parameters

start: DateLike, optional

The start date of the data to return, inclusive.

end: DateLike, optional

The end date of the data to return, inclusive.

id_type: IdType, optional

The id type to return asset ids in, e.g. ticker. The given id type must be supported by the linked universe.

Raises

ValueError

If the given id type is not supported or date range is invalid.

Returns

pl.DataFrame

The data for the given date range where the rows are the idiosyncratic volatility for each date.

abstract weights_as_task(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None) → Task[DataFrame]

abstract future_asset_returns(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None) → DataFrame

Obtains the asset returns for this risk model on the next day.

Parameters

start: DateLike, optional

The start date of the data to return, inclusive.

end: DateLike, optional

The end date of the data to return, inclusive.

id_type: IdType, optional

The id type to return asset ids in, e.g. ticker. The given id type must be supported by the linked universe.

Raises

ValueError

If the given id type is not supported or date range is invalid.

Returns

pl.DataFrame

The data for the given date range where the index is the date and the columns are the asset id. The values are the asset returns.

abstract future_asset_returns_as_task(*, start: str | date | datetime | None = None, end: str | date | datetime | None = None, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None) → Task[DataFrame]

abstract fret(*, freq: str | None = None, cumulative: bool = False, start: str | date | datetime | None = None, end: str | date | datetime | None = None) → DataFrame

Parameters

freq: str, optional

The frequency of the return aggregation, e.g. D for daily. Defaults to daily (i.e. unaggregated)

cumulative: bool, optional

If True, returns the cumulative returns.

start: DateLike, optional end: DateLike, optional

Returns

pl.DataFrame

The factor returns for the given date range.

abstract fret_as_task(*, freq: str | None = None, cumulative: bool = False, start: str | date | datetime | None = None, end: str | date | datetime | None = None) → Task[DataFrame]

abstract t_stats() → DataFrame

abstract t_stats_as_task() → Task[DataFrame]

abstract p_values() → DataFrame

abstract p_values_as_task() → Task[DataFrame]

abstract r2() → DataFrame

abstract r2_as_task() → Task[DataFrame]

abstract sigma2() → DataFrame

abstract sigma2_as_task() → Task[DataFrame]