numeraire.core.engine.PanelWeightsOutput

class numeraire.core.engine.PanelWeightsOutput(weights: Series, realized: Series, method: str, config_hash: str, data_vintage: str, run_id: str, capability: str = 'to_weights', meta: dict[str, ~typing.Any]=<factory>, missing_returns: Literal['error', 'zero', 'renormalize_legs']='error', horizon: int = 1)

Bases: object

OOS output for a cross-sectional to_weights method over a ragged panel.

weights and realized are long pd.Series on a (date, asset) MultiIndex; the wide, fixed-universe WeightsOutput can’t represent an entering/exiting universe, so the panel path carries the long form. realized is each name’s (t, t+h] return, aligned by key. weights always remains the model’s target decision; scoring_weights() separately exposes any ex-post adjustment selected through missing_returns.

__init__(weights: Series, realized: Series, method: str, config_hash: str, data_vintage: str, run_id: str, capability: str = 'to_weights', meta: dict[str, ~typing.Any]=<factory>, missing_returns: Literal['error', 'zero', 'renormalize_legs']='error', horizon: int = 1) → None

Methods

__init__(weights, realized, method, ...)
scoring_weights()	Effective ex-post weights used only to score returns under missing_returns.
strategy_returns()	Cross-sectional portfolio return per date under the missing-return policy.

Attributes

capability
horizon	Effective forecast horizon h of the paired targets (steps of the decision calendar).
missing_returns
universe	Compact universe label (n=<#assets> over the OOS panel; the name if single).
weights
realized
method
config_hash
data_vintage
run_id
meta

horizon: int = 1

Effective forecast horizon h of the paired targets (steps of the decision calendar).

Populated by backtest_panel() from the producing view; realized is each name’s (t, t+h] return, so an annualizing evaluator can distinguish overlapping multi-period targets.

property universe: str

Compact universe label (n=<#assets> over the OOS panel; the name if single).

scoring_weights() → Series

Effective ex-post weights used only to score returns under missing_returns.

strategy_returns() → Series

Cross-sectional portfolio return per date under the missing-return policy.