dfcop_dist: A discrete factor copula distribution
In tvatter/mdmd: Multivariate Discrete Mixture Distributions

Description Usage Arguments Details Value Note Examples

A S3 class to store discrete factor copula distributions.

dfcop_dist(prob, family = "indep", parameters = numeric(0))

ddfcop(x, prob, family = "indep", parameters = numeric(0), ngrid = 100)

rdfcop(n, prob, family = "indep", parameters = numeric(0))

`prob`	the marginal probabilities, a vector of numbers in (0,1).
`family`	the copula family, a string containing the family name (see Details for all possible families).
`parameters`	a vector or matrix of copula paramters.
`x`	a binary vector of the same length as prob
`ngrid`	number of nodes and weights for the Gaussian quadrature
`n`	number of observations.

The implemented families listed below. Partial matching is activated, i.e., "gauss" is equivalent to "gaussian".

indep: Independence copula.
gaussian: Gaussian copula.
t: Student t copula.
clayton: Clayton copula.
gumbel: Gumbel copula.
frank: Frank copula.
joe: Joe copula.
bb1: BB1 copula.
bb6: BB6 copula.
bb7: BB7 copula.
bb8: BB8 copula.

An object of class dfcop_dist.

The evaluation functions can optionally be used with a dfcop_dist object, e.g., ddfcop(c(1, 0), dfcop_dist(c(0.5, 0.5),"indep")).

# A 10-dimensional Gaussian factor copula
dfcop_dist(runif(10),"gaussian", 0.5)
str(dfcop_dist(runif(10), "gauss", 0.5))

# A 100-dimensional rotated Clayton factor copula
dfcop <- dfcop_dist(runif(100), "clayton", 3)
 
# evaluate the probability mass function
ddfcop(cbind(c(1,1,0,0), c(1,0,1,0)), c(0.5, 0.5), "indep")
ddfcop(rbinom(100, 1, 0.5), dfcop)

# simulate data
rdfcop(1e2, c(0.5, 0.5))
rdfcop(1e2, dfcop)