marginal

Marginal effects containers.

Classes:

Name	Description
`MeeState`	Marginal effects / estimated marginal means results.

Classes¶

MeeState¶

Marginal effects / estimated marginal means results.

Created by: build_mee_state(), dispatch_marginal_computation() Consumed by: build_effects_dataframe(), model.effects, compute_mee_inference() Augmented by: attrs.evolve() after infer() adds SEs/CIs

Attributes:

Name	Type	Description
`L_matrix`	`ndarray \| None`
`L_matrix_link`	`ndarray \| None`
`ci_lower`	`ndarray \| None`
`ci_upper`	`ndarray \| None`
`conf_level`	`float \| None`
`contrast_method`	`str \| None`
`df`	`ndarray \| None`
`effect_scale`	`str`
`estimate`	`ndarray`
`explore_formula`	`str`
`focal_var`	`str`
`grid`	`‘pl.DataFrame’`
`has_inference`	`bool`	Check if inference has been computed.
`how`	`str`
`inference_method`	`str \| None`
`link`	`str \| None`
`n_contrast_levels`	`int \| None`
`p_value`	`ndarray \| None`
`se`	`ndarray \| None`
`statistic`	`ndarray \| None`
`type`	`str`

Attributes¶

L_matrix¶

L_matrix: np.ndarray | None = field(default=None, repr=False, validator=is_optional_ndarray)

L_matrix_link¶

L_matrix_link: np.ndarray | None = field(default=None, repr=False, validator=is_optional_ndarray)

ci_lower¶

ci_lower: np.ndarray | None = field(default=None, validator=is_optional_ndarray)

ci_upper¶

ci_upper: np.ndarray | None = field(default=None, validator=is_optional_ndarray)

conf_level¶

conf_level: float | None = field(default=None, converter=normalize_optional_conf_level, validator=is_optional_conf_level)

contrast_method¶

contrast_method: str | None = field(default=None, validator=(validators.optional(validators.in_(('pairwise', 'sequential', 'poly', 'treatment', 'sum', 'helmert', 'custom')))))

df¶

df: np.ndarray | None = field(default=None, validator=is_optional_ndarray)

effect_scale¶

effect_scale: str = field(default='link', validator=(validators.in_(('link', 'response'))))

estimate¶

estimate: np.ndarray = field(validator=is_ndarray)

explore_formula¶

explore_formula: str = field(validator=(validators.instance_of(str)))

focal_var¶

focal_var: str = field(validator=(validators.instance_of(str)))

grid¶

grid: 'pl.DataFrame' = field(repr=False)

has_inference¶

has_inference: bool

Check if inference has been computed.

how¶

how: str = field(default='mem', validator=(validators.in_(('mem', 'ame'))))

inference_method¶

inference_method: str | None = field(default=None, validator=(validators.optional(validators.in_(('asymp', 'boot', 'perm')))))

link¶

link: str | None = field(default=None, validator=is_optional_str)

n_contrast_levels¶

n_contrast_levels: int | None = field(default=None, validator=is_optional_positive_int)

p_value¶

p_value: np.ndarray | None = field(default=None, validator=is_optional_ndarray)

se¶

se: np.ndarray | None = field(default=None, validator=is_optional_ndarray)

statistic¶

statistic: np.ndarray | None = field(default=None, validator=is_optional_ndarray)

type¶

type: str = field(validator=(validators.in_(('means', 'slopes', 'contrasts'))))