statsmodels.distributions.copula.api.GumbelCopula¶

class statsmodels.distributions.copula.api.GumbelCopula(theta=None, k_dim=2)[source]¶

Gumbel copula.

Dependence is greater in the positive tail than in the negative.

\[C_\theta(u,v) = \exp\!\left[ -\left( (-\log(u))^\theta + (-\log(v))^\theta \right)^{1/\theta} \right]\]

with \(\theta\in[1,\infty)\).

Methods

cdf(u[, args])	Evaluate cdf of Archimedean copula.
fit_corr_param(data)	Copula correlation parameter using Kendall's tau of sample data.
logpdf(u[, args])	Evaluate log pdf of multivariate Archimedean copula.
pdf(u[, args])	Evaluate pdf of Archimedean copula.
plot_pdf([ticks_nbr, ax])	Plot the PDF.
plot_scatter([sample, nobs, random_state, ax])	Sample the copula and plot.
rvs([nobs, args, random_state])	Draw n in the half-open interval [0, 1).
tau([theta])
tau_simulated([nobs, random_state])	Kendall's tau based on simulated samples.
theta_from_tau(tau)