prepare_anchored_subjects

prepare_anchored_subjects(
    estimates,
    as_of,
    *,
    forecast_source,
    consensus_source=CONSENSUS_SOURCE,
    lookback=_DEFAULT_LOOKBACK,
    maximum_revision_age=_DEFAULT_REVISION_AGE,
    subjects=None,
)

Anchor selected sources into stable forecast and consensus roles.

Parameters

Name	Type	Description	Default
estimates	pd.DataFrame	Long input panel; arbitrary source names are supported.	required
as_of	pd.Timestamp	Explicit UTC-aware reading moment, excluded from every read.	required
forecast_source	str	Original source to place in the forecast column.	required
consensus_source	str	Original source to place in the consensus column.	`CONSENSUS_SOURCE`
lookback	pd.Timedelta	Positive interval separating the consensus revision reads.	`_DEFAULT_LOOKBACK`
maximum_revision_age	pd.Timedelta	Positive maximum age at each revision read.	`_DEFAULT_REVISION_AGE`
subjects	pd.DataFrame \| None	Optional publication universe, including subjects with no observations. Defaults to subjects knowable strictly before as_of.	`None`

Returns

Name	Type	Description
	pd.DataFrame	One row per subject with forecast, consensus, timestamps, ages, revision and reading policy. Missing sources have typed null columns.

Raises

Name	Type	Description
	PanelError	The panel, universe, source selection or reading policy violates its contract, or a requested period end conflicts.