loss_ratio_chart

superglm.loss_ratio_chart(

y_obs,

y_pred,

sample_weight=None,

exposure=None,

*,

n_bins: int = 10,

feature_values=None,

feature_name: str | None = None,

ax: Axes | None = None,

) → LossRatioChartResult

Loss ratio chart: observed vs predicted loss ratios per bin.

Parameters:

y_obsarray-like

Observed response values.

y_predarray-like

Predicted response values.

sample_weightarray-like or None

Observation weights.

exposurearray-like or None

Exposure measure for rate models.

n_binsint

Number of quantile bins.

feature_valuesarray-like or None

If provided, bin by this feature’s values instead of predicted values.

feature_namestr or None

Label for the feature axis.

axmatplotlib Axes or None

If provided, plot onto this axes.

Returns:

LossRatioChartResult

Contains a bins DataFrame and an optional figure.