getVariancePartialRUnconditional:

In BERENZ/RISQ: What the package does (short line)

Usage

getVariancePartialRUnconditional(partialIndicator, indicator, sampleData, sampleDesign)

Arguments

partialIndicator
indicator
sampleData
sampleDesign

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

## The function is currently defined as
function (partialIndicator, indicator, sampleData, sampleDesign) 
{
    nPopulation <- sum(sampleDesign$weights)
    variable <- partialIndicator$variable
    byCategory <- partialIndicator$byCategory
    prop <- indicator$prop
    V1 <- numeric()
    V2 <- numeric()
    nSample <- nrow(sampleData)
    for (index in seq(nrow(byCategory))) {
        label <- byCategory[index, "category"]
        delta <- ifelse(sampleData[[variable]] == label, 1, 0)
        deltaC <- 1 - delta
        V1[index] <- sampleDesign$getVarTotal(sampleDesign, delta * 
            prop)
        V2[index] <- sampleDesign$getVarTotal(sampleDesign, deltaC * 
            prop)
    }
    partialIndicator$byCategory <- within(byCategory, {
        PuUnadjSE <- sqrt(n/nPopulation * (V1 * (1/n - 1/nPopulation)^2 + 
            V2 * (1/nPopulation)^2))
    })
    return(partialIndicator)
  }

BERENZ/RISQ documentation built on May 5, 2019, 10:25 a.m.