numeraire_dataset.zones.wrds_equity.MonthlyEquityTargetCollection#

class numeraire_dataset.zones.wrds_equity.MonthlyEquityTargetCollection(start_date: date, end_date: date, partitions: int, vintage: str, entries: tuple[MonthlyEquityManifestEntry, ...], manifest_hash: str, data_vintage: str, manifest_path: Path, _home: Path, _manifest_json: str)[source]#

Bases: object

SIZ monthly target returns, streamed as independent permno buckets.

__init__(start_date: date, end_date: date, partitions: int, vintage: str, entries: tuple[MonthlyEquityManifestEntry, ...], manifest_hash: str, data_vintage: str, manifest_path: Path, _home: Path, _manifest_json: str) → None#

Methods

`__init__`(start_date, end_date, partitions, ...)
`iter_clean_partitions`()	Yield no-exchange-screen SIZ returns with same-month RF, one bucket at a time.
`iter_raw_partitions`()	Yield MSF, overlapping names, and independent delists for one bucket at a time.
`manifest`()	Return a defensive copy of the frozen, path-free canonical manifest.
`raw_risk_free`()	Read the independently cached monthly Fama--French RF input.
`risk_free`()	Return decimal RF with its WRDS source-month label normalized to month-end.

Attributes

`start_date`
`end_date`
`partitions`
`vintage`
`entries`
`manifest_hash`
`data_vintage`
`manifest_path`

manifest() → dict[str, object][source]#: Return a defensive copy of the frozen, path-free canonical manifest.

iter_raw_partitions() → Iterator[MonthlyEquityRawPartition][source]#: Yield MSF, overlapping names, and independent delists for one bucket at a time.

raw_risk_free() → DataFrame[source]#: Read the independently cached monthly Fama–French RF input.

risk_free() → DataFrame[source]#: Return decimal RF with its WRDS source-month label normalized to month-end.

iter_clean_partitions() → Iterator[DataFrame][source]#

Yield no-exchange-screen SIZ returns with same-month RF, one bucket at a time.

The BAB-paper default keeps common shares but applies no exchange screen and does not invent an unreported performance-delisting return. source_ret, dlret, adjusted ret, risk_free, and excess_ret remain side-by-side for audit and downstream targets.