prior_predictive

validation.prior_predictive(
    Y0_pre,
    spatial_W,
    spatial_w,
    alpha,
    *,
    Yc_obs=None,
    X=None,
    p=0,
    a0=3.0,
    b0=1.0,
    rho_support=None,
    n_draws=2000,
    seed=123,
)

Prior predictive check of the Step-2 model.

Draws parameters from the appendix’s simple priors, forward-simulates a donor panel for each draw, computes the nine :func:ppc_stats, and – when an observed panel is supplied – reports per-statistic prior predictive p-values P(h_sim <= h_obs) over the finite simulated values.

Parameters

Name	Type	Description	Default
Y0_pre	np.ndarray	Treated pre-treatment series, shape (T0,).	required
spatial_W	np.ndarray	Raw spatial weights (normalized internally).	required
spatial_w	np.ndarray	Raw spatial weights (normalized internally).	required
alpha	np.ndarray	Fixed synthetic weights, shape (N,).	required
Yc_obs	np.ndarray	Observed donor panel (T0, N) for the observed column/p-values.	None
X	np.ndarray	Covariate cube (T0, N, K).	None
p	int	Number of latent factors.	0
a0	float	Inverse-gamma prior for sigma^2 (the replication package’s prior-predictive defaults).	3.0 / 1.0
b0	float	Inverse-gamma prior for sigma^2 (the replication package’s prior-predictive defaults).	3.0 / 1.0
rho_support	(float, float)	Support of the flat rho prior (defaults to the spectral bound).	None
n_draws	int	Number of prior draws.	2000
seed	int	Random seed.	123

Returns

Name	Type	Description
	PriorPredictiveResult