check

SuperLSS.check(

X: object | EagerFrame,

y: NDArray,

covariate: str | NDArray,

*,

name: str | None = None,

sample_weight: NDArray | None = None,

offsets: Mapping[str, NDArray] | None = None,

seed: int = 42,

**kwargs: Any,

) → BinnedCheck

Return the mean, standard deviation and skewness of the residuals per bin.

covariate is a column name of X or an array with one value per row; the rows are binned as Fasiolo, Nedellec, Goude and Wood (2020) bin them and each moment gets a bootstrap band, so a band clear of zero (mean), of one (standard deviation) or of zero (skewness) marks the region where the fit is wrong and in which moment. sample_weight is the aggregation weight of the residual payload; further keywords (n_bins, n_boot) reach the builder.