PosteriorPredictResult

Container returned by GAM.posterior_predict().

Usage

Source

PosteriorPredictResult(samples)

Holds the full (n, n_draws) posterior predictive sample matrix and provides convenience methods for computing quantile summaries, means, and intervals.

Attributes

samples: numpy.ndarray
Posterior predictive draws, shape (n, n_draws). Each column is one draw from the posterior predictive distribution (coefficient uncertainty plus observation noise).

Attributes

Name Description
n_draws Number of posterior draws.
n_obs Number of prediction points.

n_draws

Number of posterior draws.

n_draws: int


n_obs

Number of prediction points.

n_obs: int

Methods

Name Description
interval() Equal-tailed posterior predictive interval.
mean() Posterior predictive mean at each observation, shape (n,).
quantile() Compute quantile(s) of the posterior predictive distribution.
std() Posterior predictive standard deviation at each observation, shape (n,).

interval()

Equal-tailed posterior predictive interval.

Usage

Source

interval(level=0.95)

Parameters

level: float = 0.95
Coverage probability (default 0.95).

Returns

tuple[NDArray, NDArray]
(lower, upper) arrays, each of shape (n,).

mean()

Posterior predictive mean at each observation, shape (n,).

Usage

Source

mean()

quantile()

Compute quantile(s) of the posterior predictive distribution.

Usage

Source

quantile(q)

Parameters

q: float | list[float]
Quantile(s) in [0, 1]. A scalar returns shape (n,); a list returns (len(q), n).

std()

Posterior predictive standard deviation at each observation, shape (n,).

Usage

Source

std()