| CDVineCondFit {CDVineCopulaConditional} | R Documentation |
Selection of a C- or D- vine copula model for conditional sampling
Description
This function fits either a C- or a D- vine model to a d-dimensional dataset of uniform variables.
The fit of the pair-copula families is performed sequentially through the function RVineCopSelect of
the package VineCopula. The vine structure is selected among a group of C- and a D- vines which satisfy the requirement
discussed in Bevacqua et al. (2017). This group is composed by all C- and D- vines from which the conditioning variables
would be sampled as first when following the algorithms from Aas et al. (2009). Alternatively, if the
vine matrix describing the vine structure is given to the function, the fit of the pair-copulas is directly performed skipping the vine structure
selection procedure.
Usage
CDVineCondFit(data, Nx, treecrit = "AIC", type = "CVine-DVine",
selectioncrit = "AIC", familyset = NA, indeptest = FALSE,
level = 0.05, se = FALSE, rotations = TRUE, method = "mle",
Matrix = FALSE)
Arguments
data |
An |
Nx |
Number of conditioning variables. |
treecrit |
Character indicating the criteria used to select the vine. All possible vines are fitted trough
the function |
type |
Type of vine to be fitted: |
selectioncrit |
Character indicating the criterion for pair-copula selection.
Possible choices are |
familyset |
"Integer vector of pair-copula families to select from. The vector has to include at least one
pair-copula family that allows for positive and one that allows for negative
dependence. Not listed copula families might be included to better handle
limit cases. If |
indeptest |
Logical; whether a hypothesis test for the independence of
|
level |
numeric; significance level of the independence test (default:
|
se |
Logical; whether standard errors are estimated (default: |
rotations |
logical; if |
method |
indicates the estimation method: either maximum
likelihood estimation ( |
Matrix |
|
Value
An RVineMatrix object describing the selected copula model
(for further details about RVineMatrix objects see the documentation file of the VineCopula package).
The selected families are stored
in $family, and the sequentially estimated parameters in $par and $par2.
The fit of the model is performed via the function RVineCopSelect of the package VineCopula.
"The object RVineMatrix includes the following information about the fit:
se, se2standard errors for the parameter estimates (if
se = TRUE; note that these are only approximate since they do not account for the sequential nature of the estimation,nobsnumber of observations,
logLik, pair.logLiklog likelihood (overall and pairwise)
AIC, pair.AICAikaike's Informaton Criterion (overall and pairwise),
BIC, pair.BICBayesian's Informaton Criterion (overall and pairwise),
emptaumatrix of empirical values of Kendall's tau,
p.value.indeptestmatrix of p-values of the independence test.
Note
For a comprehensive summary of the vine copula model, use
summary(object); to see all its contents, use str(object)". (VineCopula Documentation, version 2.1.1, pp. 103)
Author(s)
Emanuele Bevacqua
References
Bevacqua, E., Maraun, D., Hobaek Haff, I., Widmann, M., and Vrac, M.: Multivariate statistical modelling of compound events via pair-copula constructions: analysis of floods in Ravenna (Italy), Hydrol. Earth Syst. Sci., 21, 2701-2723, https://doi.org/10.5194/hess-21-2701-2017, 2017. [link] [link]
Aas, K., Czado, C., Frigessi, A. and Bakken, H.: Pair-copula constructions of multiple dependence, Insurance: Mathematics and Economics, 44(2), 182-198, <doi:10.1016/j.insmatheco.2007.02.001>, 2009. [link]
Ulf Schepsmeier, Jakob Stoeber, Eike Christian Brechmann, Benedikt Graeler, Thomas Nagler and Tobias Erhardt (2017). VineCopula: Statistical Inference of Vine Copulas. R package version 2.1.1. [link]
See Also
Examples
# Example 1
# Read data
data(dataset)
data <- dataset$data[1:100,1:5]
# Define the variables Y and X. X are the conditioning variables,
# which have to be positioned in the last columns of the data.frame
colnames(data) <- c("Y1","Y2","X3","X4","X5")
## Not run:
# Select and fit a C- vine copula model, requiring that the
RVM <- CDVineCondFit(data,Nx=3,treecrit="BIC",type="CVine",selectioncrit="AIC")
summary(RVM)
RVM$Matrix
## End(Not run)
# Example 2
# Read data
data(dataset)
data <- dataset$data[1:80,1:5]
# Define the variables Y and X. X are the conditioning variables,
# which have to be positioned in the last columns of the data.frame
colnames(data) <- c("Y1","Y2","X3","X4","X5")
# Define a VineMatrix which can be used for conditional sampling
ListVines <- CDVineCondListMatrices(data,Nx=3)
Matrix=ListVines$DVine[[1]]
Matrix
## Not run:
# Fit copula families for the defined vine:
RVM <- CDVineCondFit(data,Nx=3,Matrix=Matrix)
summary(RVM)
RVM$Matrix
RVM$family
# check
identical(RVM$Matrix,Matrix)
# Fit copula families for the defined vine, given a group of families to select from:
RVM <- CDVineCondFit(data,Nx=3,Matrix=Matrix,familyset=c(1,2,3,14))
summary(RVM)
RVM$Matrix
RVM$family
# Try to fit copula families for a vine which is not among those
# that allow for conditional sampling:
Matrix
Matrix[which(Matrix==4)]=40
Matrix[which(Matrix==2)]=20
Matrix[which(Matrix==40)]=2
Matrix[which(Matrix==20)]=4
Matrix
RVM <- CDVineCondFit(data,Nx=3,Matrix=Matrix)
RVM
## End(Not run)