Description Usage Arguments Details Value References Examples
Select an order of the variables. The order of the variables determines the bivariate dependencies that will be explicit modeled in the first tree of the vine.
1 |
type |
Type of vine. Supported values: |
data |
Data matrix of pseudo-observations. |
method |
Ordering method. Supported values: |
... |
Additional arguments for the order method. |
In D-vines, the order of the variables determines the structure of all the trees of the vine. This is not the case for C-vines where the root node of each tree can be selected.
The "random"
method returns a random permutation of the variables.
The "greedy"
method returns an order of the variables that intends to
capture as much dependence as possible in the first tree of the vine. The
method finds the order of the variables that defines a tree that maximizes
a given dependence measure used as edge weights. For C-vines, it is determined
iteratively checking each variable as root node. For D-vines, it is equivalent
to solve the traveling salesman problem (TSP), see (Brechmann, 2010)
for details. The TSP is solved using the cheapest insertion algorithm
implemented by the solve_TSP
function of the TSP package.
The following are additional parameters for this method.
according
Dependence measure. The default value is
"kendall"
and supported values are:
"kendall"
Absolute value of Kendall's tau.
"spearman"
Absolute value of Spearman's rho.
"pearson"
Absolute value of Pearson's product moment correlation coefficient.
"df"
Smaller degrees of freedom of a bivariate t copula.
A vector with the ordered indexes of the variables. This vector should be
used to reorder the variables of the data
matrix.
Brechmann, E. C. (2010) Truncated and simplified regular vines and their applications. Diploma thesis. Technische Universitaet Muenchen.
1 2 3 4 5 |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.