weightedVar: Weighted variance and weighted standard deviation

Description

Computes a weighted variance / standard deviation of a numeric vector or across rows or columns of a matrix.

Usage

weightedVar(x, w = NULL, idxs = NULL, na.rm = FALSE, center = NULL,
  ...)
weightedSd(...)
rowWeightedVars(x, w = NULL, rows = NULL, cols = NULL, na.rm = FALSE,
  ..., useNames = TRUE)
colWeightedVars(x, w = NULL, rows = NULL, cols = NULL, na.rm = FALSE,
  ..., useNames = TRUE)
rowWeightedSds(x, w = NULL, rows = NULL, cols = NULL, na.rm = FALSE,
  ..., useNames = TRUE)
colWeightedSds(x, w = NULL, rows = NULL, cols = NULL, na.rm = FALSE,
  ..., useNames = TRUE)

Value

Returns a numeric scalar.

Arguments

x: vector of type integer, numeric, or logical.
w: a vector of weights the same length as x giving the weights to use for each element of x. Negative weights are treated as zero weights. Default value is equal weight to all values.
idxs: A vector indicating subset of elements to operate over. If NULL, no subsetting is done.
na.rm: If TRUE, missing values are excluded.
center: Optional numeric scalar specifying the center location of the data. If NULL, it is estimated from data.
...: Not used.
rows: A vector indicating subset of rows to operate over. If NULL, no subsetting is done.
cols: A vector indicating subset of columns to operate over. If NULL, no subsetting is done.
useNames: If FALSE (default), no naming support is done. Else if TRUE, names attributes of result are set.

Missing values

This function handles missing values consistently with weightedMean(). More precisely, if na.rm = FALSE, then any missing values in either x or w will give result NA_real_. If na.rm = TRUE, then all (x, w) data points for which x is missing are skipped. Note that if both x and w are missing for a data points, then it is also skipped (by the same rule). However, if only w is missing, then the final results will always be NA_real_ regardless of na.rm.

Author

Henrik Bengtsson

Details

The estimator used here is the same as the one used by the "unbiased" estimator of the Hmisc package. More specifically, weightedVar(x, w = w) == Hmisc::wtd.var(x, weights = w),