Skip to contents

Weighted Statistical Functions for Posterior Analysis

Source: R/weighted_statistics.R
weighted_var.Rd

Helper functions for calculating weighted statistics used in posterior parameter distribution analysis. Weighted variance calculation

Usage

weighted_var(x, w)

Arguments

x

Numeric vector of values

w

Numeric vector of weights (same length as x)

Value

Weighted variance (scalar)

Details

Calculates the bias-corrected weighted variance for a vector of values.

Uses the bias correction: Var = sum(w * (x - mu)^2) / (sum(w) - sum(w^2)/sum(w)) where mu is the weighted mean. Returns 0 for single values or invalid denominators.

Examples

x <- c(1, 2, 3, 4, 5)
w <- c(0.1, 0.2, 0.4, 0.2, 0.1)
weighted_var(x, w)
#> [1] 1.621622