Combine a fitted bivariate (copula) analysis with an M x N response surface Z = f(X, Y) to
derive the annual-exceedance-probability curve of Z via the conditional-frequency law of total
probability. Wraps the shared C++ CoincidentFrequencyAnalysis (which internally fits the
bivariate analysis from the same marginal/copula spec).
Usage
coincident_frequency_analysis(
marginal_x_family,
marginal_x_data,
marginal_x_parameters,
marginal_y_family,
marginal_y_data,
marginal_y_parameters,
x_values,
y_values,
response,
number_of_bins = 50L,
copula = "Normal",
estimation_method = "InferenceFromMargins",
sampler = "DEMCz",
iterations = 3000L,
output_length = 10000L,
credible_level = 0.9,
seed = 12345L,
number_of_chains = 4L,
thinning_interval = -1L
)Arguments
- marginal_x_family, marginal_y_family
marginal distribution family names.
- marginal_x_data, marginal_y_data
numeric vectors of the marginal samples.
- marginal_x_parameters, marginal_y_parameters
numeric fixed marginal parameter vectors.
- x_values, y_values
numeric ascending vectors of the primary (X) / secondary (Y) ordinates.
- response
numeric M x N matrix of the response surface
Z[i, j] = f(x_i, y_j).- number_of_bins
number of Z output bins (default
50).- copula
copula family name (e.g.
"Normal","Clayton","Gumbel").- estimation_method
copula estimation method (default
"InferenceFromMargins").- sampler
MCMC sampler:
"DEMCz"(default),"DEMCzs","ARWMH", or"NUTS".- iterations
number of post-warmup MCMC iterations.
- output_length
number of posterior samples used to build the credible band.
- credible_level
credible-interval width (e.g.
0.90for a 90% band).- seed
PRNG seed for the sampler (fixed for reproducibility).
- number_of_chains
number of MCMC chains (default
4).- thinning_interval
MCMC thinning interval;
-1(default) keeps the sampler's own default.