Weighted sample quantiles

Share:

Description

Compute quantiles taking into account sample weights. The following methods are implemented:

  • quantileWt.default(x, weights=NULL, probs=seq(0, 1, 0.25), na.rm=TRUE, ...)

  • quantileWt.dataObj(x, vars, probs=seq(0, 1, 0.25), na.rm=TRUE, ...)

Additional parameters are:

  • weights an optional numeric vector containing sample weights.

  • vars a character vector of length 1 specifying a variable name that is available in the data-slot of x and which is used for the calculation.

  • probs a numeric vector of probabilities with values in [0, 1].

  • na.rm a logical indicating whether any NA or NaN values should be removed from x before the quantiles are computed. Note that the default is TRUE, contrary to the function quantile.

Usage

1

Arguments

x

a numeric vector.

...

for the generic function quantileWt additional arguments to be passed to methods. Additional arguments not included in the definition of the methods are currently ignored.

Details

If weights are not specified then quantile(x, probs, na.rm=na.rm, names=FALSE, type=1) is used for the computation.

Note probabilities outside [0, 1] cause an error.

Value

A vector of the (weighted) sample quantiles.

Author(s)

Stefan Kraft and Bernhard Meindl

A basic version of this function was provided by Cedric Beguin and Beat Hulliger.

See Also

quantile

Examples

1
2
3
4
5
6
data(eusilcS)
(quantileWt(eusilcS$netIncome, weights=eusilcS$rb050))

# dataObj-method
inp <- specifyInput(data=eusilcS, hhid="db030", hhsize="hsize", strata="db040", weight="db090")
(quantileWt(inp, vars="netIncome"))

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.