Fit a copula by maximum-likelihood via copula — morie_copula

Thin extender over copula::fitCopula that estimates copula parameters from pseudo-observations on $[0, 1]^d$.

Usage

morie_copula_fit(copula, data, ...)

Arguments

copula: A copula object specifying the parametric family (e.g. copula::normalCopula(), copula::claytonCopula()).
data: Numeric matrix of pseudo-observations on $[0, 1]^d$, with one column per margin (typically obtained via copula::pobs).
...: Further arguments forwarded to copula::fitCopula (e.g. method, start, optim.method, estimate.variance).

Value

A list with $method = "copula::fitCopula" and $raw (a fitCopula object with the estimated parameters, log-likelihood and variance estimates).

Examples

# \donttest{
  if (requireNamespace("copula", quietly = TRUE)) {
    set.seed(1)
    cop <- copula::normalCopula(0.5, dim = 2)
    u <- copula::rCopula(200, cop)
    morie_copula_fit(copula::normalCopula(dim = 2), data = u)
  }
#> $method
#> [1] "copula::fitCopula"
#> 
#> $raw
#> Call: copula::fitCopula(copula, data = data, ...)
#> Fit based on "maximum pseudo-likelihood" and 200 2-dimensional observations.
#> Copula: normalCopula 
#>  rho.1 
#> 0.5106 
#> The maximized loglikelihood is 27.41 
#> Optimization converged
#> 
# }