bayesline.api.equity.UniverseApi

class bayesline.api.equity.UniverseApi

Bases: ABC

Methods

__init__()
counts([dates, industry_level, ...])	Parameters dates: bool, optional If True, groups by dates industry_level: int, optional The level of industry aggregation to group by. 0 means no industry aggregation, 1 means level 1, etc. Values greater than the max level are treated as the max level. region_level: int, optional The level of region aggregation to group by. 0 means no region aggregation, 1 means level 1, etc. Values greater than the max level are treated as the max level. universe_type: Literal["estimation", "coverage", "both"], optional The type of universe to calculate the counts for. id_type: IdType, optional The id type to calculate the daily stats for, e.g. ticker, which is relevant as the coverage may differ by id type. The given id type must be supported, i.e. in id_types. labels: bool, optional If True, return labels for the counts, otherwise use the codes.
counts_as_task([dates, industry_level, ...])
coverage([id_type])	Parameters id_type: IdType, optional The id type to return asset ids in, e.g. ticker
dates(*[, range_only, trade_only])	Parameters range_only: bool, default=False If True, returns the first and last date only. trade_only: bool, default=False If True, filter down the dats to trade dates only.
get(*[, start, end, id_type, filter_tradedays])	Parameters start: DateLike, optional The start date of the universe 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, i.e. in id_types. filter_tradedays: bool, default=False If True, filter down the data to trade dates only.
get_as_task(*[, start, end, id_type, ...])
input_id_mapping(*[, id_type, filter_mode, mode])	Parameters id_type: IdType, optional The id type to return asset ids in, e.g. ticker, or the default ID type of the universe if None. filter_mode: Literal[all, mapped, unmapped] if mapped will only consider assets that could be mapped. if unmapped will only consider assets that could not be mapped. mode: Literal[all, daily-counts, latest-name] if all, returns all dated mappings if daily-counts, returns the daily counts of mapped assets if input-asset-counts, returns the total counts of input assets if latest-name, returns the latest name of mapped assets

Attributes

id_types	Returns supported id types for this universe.
settings	Returns The settings used to create this universe.

abstract property settings: UniverseSettings

Returns

The settings used to create this universe.

abstract property id_types: list[Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name']]

Returns

supported id types for this universe.

abstract coverage(id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None) → list[str]

Parameters

id_type: IdType, optional

The id type to return asset ids in, e.g. ticker

Raises

ValueError

If the given id type is not supported.

Returns

list of all asset ids this universe covers, in given id type.

abstract dates(*, range_only: bool = False, trade_only: bool = False) → list[date]

Parameters

range_only: bool, default=False

If True, returns the first and last date only.

trade_only: bool, default=False

If True, filter down the dats to trade dates only.

Returns

list of all dates this universe covers.

abstract counts(dates: bool = True, industry_level: int = 0, region_level: int = 0, universe_type: Literal['estimation', 'coverage', 'both'] = 'both', id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None, labels: bool = True) → DataFrame

Parameters

dates: bool, optional

If True, groups by dates

industry_level: int, optional

The level of industry aggregation to group by. 0 means no industry aggregation, 1 means level 1, etc. Values greater than the max level are treated as the max level.

region_level: int, optional

The level of region aggregation to group by. 0 means no region aggregation, 1 means level 1, etc. Values greater than the max level are treated as the max level.

universe_type: Literal[“estimation”, “coverage”, “both”], optional

The type of universe to calculate the counts for.

id_type: IdType, optional

The id type to calculate the daily stats for, e.g. ticker, which is relevant as the coverage may differ by id type. The given id type must be supported, i.e. in id_types.

labels: bool, optional

If True, return labels for the counts, otherwise use the codes.

Returns

pl.DataFrame

Universe counts. If grouped by dates then the count will be given. If not grouped by dates then the mean/min/max across all dates will be given.

abstract counts_as_task(dates: bool = True, industry_level: int = 0, region_level: int = 0, universe_type: Literal['estimation', 'coverage', 'both'] = 'both', id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None, labels: bool = True) → Task[DataFrame]

abstract input_id_mapping(*, id_type: Literal['bayesid', 'ticker', 'composite_figi', 'cik', 'cusip8', 'cusip9', 'isin', 'sedol6', 'sedol7', 'proxy', 'name'] | None = None, filter_mode: Literal['all', 'mapped', 'unmapped'] = 'all', mode: Literal['all', 'daily-counts', 'input-asset-counts', 'latest-name'] = 'all') → DataFrame

Parameters

id_type: IdType, optional

The id type to return asset ids in, e.g. ticker, or the default ID type of the universe if None.

filter_mode: Literal[all, mapped, unmapped]

if mapped will only consider assets that could be mapped. if unmapped will only consider assets that could not be mapped.

mode: Literal[all, daily-counts, latest-name]

if all, returns all dated mappings if daily-counts, returns the daily counts of mapped assets if input-asset-counts, returns the total counts of input assets if latest-name, returns the latest name of mapped assets

Returns

pl.DataFrame

If mode is all, a DataFrame with date, input_asset_id, input_asset_id_type, output_asset_id, output_asset_id_type and, name columns. It contains contains the original input ID space and the mapped ids. The mapped IDs will be None if for the given date and input ID the asset cannot be mapped. If mode is daily-counts, a DataFrame with date and count columns. If mode is input-asset-counts, a DataFrame with input_asset_id and count columns. If mode is latest-name, a DataFrame with asset_id and name columns.

abstract get(*, 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, filter_tradedays: bool = False) → DataFrame

Parameters

start: DateLike, optional

The start date of the universe 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, i.e. in id_types.

filter_tradedays: bool, default=False

If True, filter down the data to trade dates only.

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.

abstract get_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, filter_tradedays: bool = False) → Task[DataFrame]