estimation_diagnostics

estimation_diagnostics(
    data,
    distribution,
    sampler='DEMCz',
    iterations=3000,
    output_length=10000,
    seed=12345,
    thinning_interval=-1,
    thin_every=10,
)

Bayesian estimation diagnostics (leverage / influence / prior influence).

Fit distribution to data with a Bayesian MCMC analysis and compute the three diagnostics off that fit.

Parameters

Name	Type	Description	Default
data	array_like	Observations to fit.	required
distribution	str	Distribution family name.	required
sampler	('DEMCz', 'DEMCzs', 'ARWMH', 'NUTS')	MCMC sampler.	"DEMCz"
iterations	int	Number of post-warmup MCMC iterations.	3000
output_length	int	Number of posterior samples.	10000
seed	int	PRNG seed for the sampler.	12345
thinning_interval	int	MCMC thinning interval; -1 keeps the sampler’s own default.	-1
thin_every	int	Prior-influence posterior thinning stride.	10

Returns

Name	Type	Description
	dict	Three sub-dicts: - leverage : the lists index, leverage, fit_influence, variance_influence, value; prior_leverage; and total_leverage, total_fit_influence, total_variance_influence. - influence : the lists pareto_k, elpd_loo; and count, mean_pareto_k, max_pareto_k, count_pareto_k_above_05 / _07 / _10, proportion_problematic, is_reliable. - prior_influence : count, prior_precision_share, total_prior_log_likelihood, total_data_log_likelihood, prior_to_data_ratio, is_prior_influential, mean_prior_precision_share.