numeraire.core.evaluators.CrossSectionalR2Evaluator

class numeraire.core.evaluators.CrossSectionalR2Evaluator

Bases: object

Cross-sectional R^2 of mean realized returns on mean predicted expected returns (OLS).

The pricing headline (the classic average-realized-vs-average-predicted plot): time-average each asset’s realized and predicted returns, then OLS-regress mean realized on mean predicted across assets and report the R^2. Assets missing either mean are dropped. Read against the output’s protocol — an "in_sample" R^2 is explanatory, a "walk_forward" R^2 is out-of-sample.

__init__()

Methods

__init__()
evaluate(oos_output)

Attributes

requires