pacotestRvineSingleCopula {pacotest} | R Documentation |
Testing for a Single (j-1)-th Order Partial Copula in a R-Vine Copula
Description
The function can be used to test a single copula in a R-vine copula to be a (j-1)-th order partial copula.
To apply the function one needs to provide the data and a specified/estimated R-vine copula model in form of a RVineMatrix from the VineCopula-package.
Additionally, a pacotest options list, which can be generated with the pacotestset
function, needs to be provided.
Usage
pacotestRvineSingleCopula(data, RVM, pacotestOptions, tree, copulaNumber)
Arguments
data |
A (n x d) matrix (or data frame) of [0,1] data (i.e. uniform margins). |
RVM |
An RVineMatrix object (VineCopula-package) which includes the structure, the pair-copula families and parameters of an R-vine copula. |
pacotestOptions |
A options list generated by the |
tree |
The tree number (j>=2) of the copula which should be tested to be a (j-1)-th order partial copula. |
copulaNumber |
The number (1<= copulaNumber <= j-1) of the copula in the normalized RVineMatrix which should be tested to be a (j-1)-th order partial copula. |
Value
A list which can, depending on the chosen test, consist of the following elements:
pValue |
The p-value of the test. |
testStat |
The value of the test statistic. |
decisionTree |
The decision tree used to partition the support Lmabda0 of the conditioning variable W. It is provided as a list consisting of three nodes ( |
S |
The bootstrapped values of the test statistic (only for the test type |
Author(s)
Malte S. Kurz
References
Kurz, M. S. and F. Spanhel (2022), "Testing the simplifying assumption in high-dimensional vine copulas", Electronic Journal of Statistics 16 (2), pp. 5226-5276.
Spanhel, F. and M. S. Kurz (2019), "Simplified vine copula models: Approximations based on the simplifying assumption", Electronic Journal of Statistics 13 (1), pp. 1254-1291.
See Also
pacotest-package
, pacotest
, pacotestset
, pacotestRvineSeq
Examples
# Sample data and R-vine copula selection are taken
# from the documentation of RVineStructureSelect
# of the VineCopula package.
# Obtain sample data
data(daxreturns, package ="VineCopula")
dataSet = daxreturns[1:750,1:4]
# Specify an R-vine copula model
# (can be obtained by calling: RVM = VineCopula::RVineStructureSelect(dataSet))
vineStructure = matrix(c(3,4,1,2,0,2,4,1,0,0,1,4,0,0,0,4),4,4)
families = matrix(c(0,5,2,2,0,0,2,14,0,0,0,14,0,0,0,0),4,4)
par = matrix(c(0,0.8230664,0.1933472,0.6275062,
0,0,0.2350109,1.6619945,
0,0,0,1.599363,
0,0,0,0),4,4)
par2 = matrix(c(0,0,11.757700,4.547847,
0,0,17.15717,0,
0,0,0,0,0,0,0,0),4,4)
RVM = VineCopula::RVineMatrix(vineStructure, families, par, par2)
# Specify a pacotestOptions list:
# For illustrating the functioning of the decision tree,
# grouped scatterplots and a decision tree plot are activated.
pacotestOptions = pacotestset(testType='CCC',
groupedScatterplots = TRUE,
decisionTreePlot = TRUE)
# Test for a 2-nd order partial copula
# corresponding to the variables BAYN.DE,BMW.DE
# and conditioning set ALV.DE,BAS.DE
tree = 3
copulaNumber = 1
pacotestResultList = pacotestRvineSingleCopula(dataSet, RVM,
pacotestOptions, tree, copulaNumber)