Estimation of the Variance-Gamma Distribution

Performs maximum likelihood estimation on the Variance-Gamma (VG) distribution (univariate and multivariate). Wrappers fit.VGuv and fit.VGmv.

fit_vg(x, symmetric = FALSE)

# S3 method for default
fit_vg(x, symmetric = FALSE)

# S3 method for tbl
fit_vg(x, symmetric = FALSE)

# S3 method for xts
fit_vg(x, symmetric = FALSE)

# S3 method for matrix
fit_vg(x, symmetric = FALSE)

Arguments

x	A tabular (non-tidy) data structure.
symmetric	A logical flag. Should the estimated distribution be symmetric? Defaults to FALSE.

Value

A list of the the class cma_fit with 21 components.

See also

fit_ghd fit_hyp fit_nig fit_t fit_normal

Examples

x <- matrix(diff(log(EuStockMarkets)), ncol = 4)

# multivariate estimation
fit_vg(x[ , 3:4])
#> # Margins Estimation
#> Converged:       TRUE
#> Dimension:       2
#> AIC:            -25416.89
#> Log-Likelihood:  12716.45
#> Model:           Asymmetric Variance Gamma

# univariate estimation
fit_vg(x[ , 4, drop = FALSE])
#> Singularity (x-mu)==0: Interpolate with splines.
#> Singularity (x-mu)==0: Interpolate with splines.
#> Singularity (x-mu)==0: Interpolate with splines.
#> Singularity (x-mu)==0: Interpolate with splines.
#> # Margins Estimation
#> Converged:       TRUE
#> Dimension:       1
#> AIC:            -12783.01
#> Log-Likelihood:  6395.503
#> Model:           Asymmetric Variance Gamma