spatial_gev_analysis

spatial_gev_analysis(
    coordinates,
    at_site_data,
    cross_validation=False,
    sampler='DEMCz',
    iterations=3000,
    output_length=10000,
    credible_level=0.9,
    seed=12345,
    number_of_chains=4,
    exceedance_probabilities=None,
    thinning_interval=-1,
)

Hierarchical spatial-GEV frequency analysis over gauged sites.

Wraps the shared C++ SpatialGEVAnalysis.

Parameters

Name	Type	Description	Default
coordinates	array_like	Site coordinates, one [x, y] row per site.	required
at_site_data	array_like	At-site observation matrix, [observations x sites].	required
cross_validation	bool	Also compute the leave-one-site-out diagnostics.	False
sampler	('DEMCz', 'DEMCzs', 'ARWMH', 'NUTS')	MCMC sampler.	"DEMCz"
iterations	int	Number of post-warmup MCMC iterations.	3000
output_length	int	Number of posterior samples used to build the credible bands.	10000
credible_level	float	Credible-interval width.	0.90
seed	int	PRNG seed for the sampler.	12345
number_of_chains	int	Number of MCMC chains.	4
exceedance_probabilities	array_like of float	Exceedance probabilities at which to tabulate the curves; when None, the default ordinates are used.	None
thinning_interval	int	MCMC thinning interval; -1 keeps the sampler’s own default.	-1

Returns

Name	Type	Description
	dict	The regional frequency curve plus per-site GEV/quantile bands and, when cross_validation is true, the leave-one-site-out diagnostics.