Description Usage Value Author(s)
n = #replications or sample size pmat = nxd matrix of levels of u1, C_2|1(u_2|u_1),..., C_d|1..d-1(u_d|u[1:(d-1)]) parvec = vector of parameters to be optimized in nllk A = dxd vine array with 1:d on diagonal ntrunc = truncated level, assume >=1 pcondmat = matrix of names of conditional cdfs for trees 1,...,ntrunc (assuming only one needed for permutation symmetric pair-copulas) qcondmat = matrix of names of conditional quantile functions for trees 1,...,ntrunc qcondmat and pcondmat are empty for diagonal and lower triangle, and could have ntrunc rows or be empty for rows ntrunc+1 to d-1 np = dxd where np[ell,j] is #parameters for parameter th[ell,j] for pair-copula in tree ell, variables j and A[ell,j] np=0 on and below diagonal iinv=T to check that this is inverse of rvinepcond() in this case, columns of pmat come from rvinepcond() iinv=F, get quantiles C_d|1:(d-1)(p|u[1:(d-1)]) based on last column of pmat[,d] where u[1],..,u[d-1] have been previously converted to u[1], C_2|1(u[2]|u[1]), ... C_d-1|1:(d-2)(u[d-1]|u[1:(d-2)]) via rvinepcond()
1 | rvineqcond(pmat, A, ntrunc, parvec, np, qcondmat, pcondmat, iinv = F)
|
n x d matrix with values in (0,1) or quantile C_d|1...d-1(p| u[1:(d-1)]) [This function is modification of rvinesimvec2 in CopulaModel]
Harry Joe
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.