Automated fitting or creation of custom vine copula models
vine(
data,
margins_controls = list(mult = NULL, xmin = NaN, xmax = NaN, bw = NA, deg = 2),
copula_controls = list(family_set = "all", structure = NA, par_method = "mle",
nonpar_method = "constant", mult = 1, selcrit = "aic", psi0 = 0.9, presel = TRUE,
trunc_lvl = Inf, tree_crit = "tau", threshold = 0, keep_data = FALSE, show_trace =
FALSE, cores = 1),
weights = numeric(),
keep_data = FALSE,
cores = 1
)
vine_dist(margins, pair_copulas, structure)a matrix or data.frame. Discrete variables have to be declared as
ordered().
a list with arguments to be passed to
kde1d::kde1d(). Currently, there can be
mult numeric vector of length one or d; all bandwidths for marginal
kernel density estimation are multiplied with mult. Defaults to
log(1 + d) where d is the number of variables after applying
rvinecopulib:::expand_factors().
xmin numeric vector of length d; see kde1d::kde1d().
xmax numeric vector of length d; see kde1d::kde1d().
bw numeric vector of length d; see kde1d::kde1d().
deg numeric vector of length one or d; kde1d::kde1d().
a list with arguments to be passed to vinecop().
optional vector of weights for each observation.
whether the original data should be stored; if you want to
store the pseudo-observations used for fitting the copula, use the
copula_controls argument.
the number of cores to use for parallel computations.
A list with with each element containing the specification of a
marginal stats::Distributions. Each marginal specification
should be a list with containing at least the distribution family ("distr")
and optionally the parameters, e.g.
list(list(distr = "norm"), list(distr = "norm", mu = 1), list(distr = "beta", shape1 = 1, shape2 = 1)).
Note that parameters that have no default values have to be provided.
Furthermore, if margins has length one, it will be recycled for every component.
A nested list of 'bicop_dist' objects, where
pair_copulas[[t]][[e]] corresponds to the pair-copula at edge e in
tree t.
an rvine_structure object, namely a compressed
representation of the vine structure, or an object that can be coerced
into one (see rvine_structure() and as_rvine_structure()).
The dimension must be length(pair_copulas[[1]]) + 1.
Objects inheriting from vine_dist for vine_dist(), and
vine and vine_dist for vine().
Objects from the vine_dist class are lists containing:
margins, a list of marginals (see below).
copula, an object of the class vinecop_dist, see vinecop_dist().
For objects from the vine class, copula is also an object of the class
vine, see vinecop(). Additionally, objects from the vine class contain:
margins_controls, a list with the set of fit controls that was passed
to kde1d::kde1d() when estimating the margins.
copula_controls, a list with the set of fit controls that was passed
to vinecop() when estimating the copula.
data (optionally, if keep_data = TRUE was used), the dataset that was
passed to vine().
nobs, an integer containing the number of observations that was used
to fit the model.
Concerning margins:
For objects created with vine_dist(), it simply corresponds to the margins
argument.
For objects created with vine(), it is a list of objects of class kde1d,
see kde1d::kde1d().
vine_dist() creates a vine copula by specifying the margins, a nested list
of bicop_dist objects and a quadratic structure matrix.
vine() provides automated fitting for vine copula models.
margins_controls is a list with the same parameters as
kde1d::kde1d() (except for x). copula_controls is a list
with the same parameters as vinecop() (except for data).
# specify pair-copulas
bicop <- bicop_dist("bb1", 90, c(3, 2))
pcs <- list(
list(bicop, bicop), # pair-copulas in first tree
list(bicop) # pair-copulas in second tree
)
# specify R-vine matrix
mat <- matrix(c(1, 2, 3, 1, 2, 0, 1, 0, 0), 3, 3)
# set up vine copula model with Gaussian margins
vc <- vine_dist(list(distr = "norm"), pcs, mat)
# show model
summary(vc)
#> $margins
#> # A data.frame: 3 x 2
#> margin distr
#> 1 norm
#> 2 norm
#> 3 norm
#>
#> $copula
#> # A data.frame: 3 x 10
#> tree edge conditioned conditioning var_types family rotation parameters df
#> 1 1 3, 1 c,c bb1 90 3, 2 2
#> 1 2 2, 1 c,c bb1 90 3, 2 2
#> 2 1 3, 2 1 c,c bb1 90 3, 2 2
#> tau
#> -0.8
#> -0.8
#> -0.8
#>
# simulate some data
x <- rvine(50, vc)
# estimate a vine copula model
fit <- vine(x, copula_controls = list(family_set = "par"))
summary(fit)
#> $margins
#> # A data.frame: 3 x 6
#> margin name nobs bw loglik d.f.
#> 1 V1 50 1.6 -81 3.6
#> 2 V2 50 1.8 -81 5.1
#> 3 V3 50 1.2 -79 2.3
#>
#> $copula
#> # A data.frame: 3 x 11
#> tree edge conditioned conditioning var_types family rotation parameters df
#> 1 1 2, 1 c,c gumbel 270 8 1
#> 1 2 1, 3 c,c bb1 270 2.1, 3.7 2
#> 2 1 2, 3 1 c,c gumbel 90 4.1 1
#> tau loglik
#> -0.88 83
#> -0.87 82
#> -0.75 51
#>
## model for discrete data
x <- as.data.frame(x)
x[, 1] <- ordered(round(x[, 1]), levels = seq.int(-5, 5))
fit_disc <- vine(x, copula_controls = list(family_set = "par"))
summary(fit_disc)
#> $margins
#> # A data.frame: 3 x 6
#> margin name nobs bw loglik d.f.
#> 1 V1 50 1.6 -81 1.3
#> 2 V2 50 1.8 -81 5.1
#> 3 V3 50 1.2 -79 2.3
#>
#> $copula
#> # A data.frame: 3 x 11
#> tree edge conditioned conditioning var_types family rotation parameters df
#> 1 1 2, 1 c,d gumbel 270 8.6 1
#> 1 2 1, 3 d,c bb7 270 5.4, 23.2 2
#> 2 1 2, 3 1 c,c frank 0 1.6 1
#> tau loglik
#> -0.88 59.1
#> -0.86 60.0
#> 0.18 2.4
#>