term_inference#

SuperGLM.term_inference(
name: Hashable,
*,
with_se: bool = True,
simultaneous: bool = False,
n_points: int = 200,
alpha: float = 0.05,
n_sim: int = 10000,
seed: int = 42,
centering: str = 'native',
) → TermInference | InteractionInference#

Per-term inference: curve, uncertainty, and metadata in one object.

Parameters:
centering{“native”, “mean”}

"native" (default) returns the canonical fitted term contribution under the model’s identifiability constraint. "mean" is a reporting convenience that shifts so the geometric mean of relativities = 1. It is a change of identifiability constraint, so the standard errors and intervals change with it: they become those of the centered contrast, and the level the fit pinned stops carrying a zero-width interval.