R/utilitiesRVineEqualCorrWithEstimationUncertainty.R

Defines functions testStatEqualCorrWithoutEstimation testStatEqualCorrWithEstimation covOfCovariancesWithEstimation covOfCorrelationsWithEstimation

covOfCorrelationsWithEstimation = function(data, svcmDataFrame, indList, cPitData, theta, estUncertWithRanks = FALSE)
{
  
  gInv = gInvRvine(data, svcmDataFrame, indList, cPitData, theta)
  omega = omegaRvine(data, svcmDataFrame, indList, cPitData, theta, estUncertWithRanks)
  
  varMat = gInv %*% omega %*% t(gInv)
  
  return(varMat)
}

covOfCovariancesWithEstimation = function(data, svcmDataFrame, indList, cPitData, theta, estUncertWithRanks = FALSE)
{
  
  gInv = gInvRvineCov(data, svcmDataFrame, indList, cPitData, theta)
  omega = omegaRvineCov(data, svcmDataFrame, indList, cPitData, theta, estUncertWithRanks)
  
  varMat = gInv %*% omega %*% t(gInv)
  
  return(varMat)
}


testStatEqualCorrWithEstimation = function(data, svcmDataFrame, ind)
{
  d <- ncol(data)
  # Check whether the transferred indicator matrix is a valid grouping indicator matrix
  ind = as.matrix(ind)
  nGroups = ncol(ind)
  if (nGroups < 2 ) 
    stop("At least two groups have to be specified.")
  if (nGroups > 4 ) 
    stop("The maximum number of groups is 4.")
  if (dim(data)[1] != dim(ind)[1]) 
    stop("Dimensions of 'data' and 'ind' do not match.")
  if (any(rowSums(ind)>1))
    stop("Non disjunct groups are not implemented.")
  
  ind = (ind == 1) # transfer the possibly numeric matrix into a matrix of logicals
  
  
  nGroups = ncol(ind)
  nObs = nrow(data)
  
  indexVectors = matrix(0, nrow = nObs, ncol = nGroups)
  nObsPerVector = vector(length = nGroups)
  
  inds = seq(from = 0,to = (nObs-1), by = 1)
  
  for (iGroup in 1:nGroups)
  {
    xx = inds[ind[,iGroup]]
    nObsPerVector[iGroup] = length(xx)
    
    indexVectors[1:nObsPerVector[iGroup],iGroup] = xx
  }
  
  
  
  # Compute CPITs for the whole vine
  cPitData = getCpitsFromVine(data, svcmDataFrame)
  
  # Obtain the cPits to be tested
  copulaInd = nrow(svcmDataFrame)
  cPit1 = getCpit1(cPitData, svcmDataFrame, copulaInd)
  cPit2 = getCpit2(cPitData, svcmDataFrame, copulaInd)
  
  out = testStatEqualCorrWithEstimationCpp(indexVectors, nObsPerVector, cbind(cPit1,cPit2), data, svcmDataFrame, cPitData)
  
  out$pValue = 1 - pchisq(out$testStat,nGroups-1)
  
  return(out)
  
}


testStatEqualCorrWithoutEstimation = function(data, svcmDataFrame, ind)
{
  d <- ncol(data)
  # Check whether the transferred indicator matrix is a valid grouping indicator matrix
  ind = as.matrix(ind)
  nGroups = ncol(ind)
  if (nGroups < 2 ) 
    stop("At least two groups have to be specified.")
  if (nGroups > 4 ) 
    stop("The maximum number of groups is 4.")
  if (dim(data)[1] != dim(ind)[1]) 
    stop("Dimensions of 'data' and 'ind' do not match.")
  if (any(rowSums(ind)>1))
    stop("Non disjunct groups are not implemented.")
  
  
  nGroups = ncol(ind)
  nObs = nrow(data)
  
  indexVectors = matrix(0, nrow = nObs, ncol = nGroups)
  nObsPerVector = vector(length = nGroups)
  
  inds = seq(from = 0,to = (nObs-1), by = 1)
  
  for (iGroup in 1:nGroups)
  {
    xx = inds[ind[,iGroup]]
    nObsPerVector[iGroup] = length(xx)
    
    indexVectors[1:nObsPerVector[iGroup],iGroup] = xx
  }
  
  # Compute CPITs for the whole vine
  cPitData = getCpitsFromVine(data, svcmDataFrame)
  
  # Obtain the cPits to be tested
  copulaInd = nrow(svcmDataFrame)
  cPit1 = getCpit1(cPitData, svcmDataFrame, copulaInd)
  cPit2 = getCpit2(cPitData, svcmDataFrame, copulaInd)
  
  out = testStatEqualCorrWithoutEstimationCpp(indexVectors, nObsPerVector, cbind(cPit1,cPit2))
  
  out$pValue = 1 - pchisq(out$testStat,nGroups-1)
  
  return(out)
}

Try the pacotest package in your browser

Any scripts or data that you put into this service are public.

pacotest documentation built on May 2, 2019, 6:52 a.m.