varW: Weighted Variance

In murphymv/semEff: Automatic Calculation of Effects for Piecewise Structural Equation Models

Weighted Variance

Description

Calculate the weighted variance of x.

Usage

varW(x, w = NULL, na.rm = FALSE)

Arguments

x	A numeric vector.
w	A numeric vector of weights of the same length as x.
na.rm	Logical, whether NAs in x should be removed.

Details

Calculate the weighted variance of x via the weighted covariance matrix (cov.wt()). If no weights are supplied, the simple variance is returned instead. As in weighted.mean(), NAs in w are not handled specially and will return NA as result.

Value

A numeric value, the weighted variance of x.

See Also

var()

Examples

# Weighted variance
x <- rnorm(30)
w <- runif(30, 0, 1)
varW(x, w)

# Simple variance
varW(x)
stopifnot(varW(x) == var(x))

# NA handling
varW(c(x[1:29], NA), w, na.rm = TRUE)  # NA in x (removed)
varW(c(x[1:29], NA), w, na.rm = FALSE)  # NA in x (NA returned)
varW(x[1:29], w = c(w[1:29], NA))  # NA in w (NA returned)

murphymv/semEff documentation built on Sept. 18, 2024, 7:40 p.m.