wtd.mean: Weighted Arithmetic Mean

In npreg: Nonparametric Regression via Smoothing Splines

Weighted Arithmetic Mean

Description

Generic function for calculating the weighted (and possibly trimmed) arithmetic mean.

Usage

wtd.mean(x, weights, trim = 0, na.rm = FALSE)

Arguments

x	Numerical or logical vector.
weights	Vector of non-negative weights.
trim	Fraction [0, 0.5) of observations trimmed from each end before calculating mean.
na.rm	Logical indicating whether NA values should be removed before calculation.

Details

If weights are missing, the weights are defined to be a vector of ones (which is the same as the unweighted arithmetic mean).

If trim is non-zero, then trim observations are deleted from each end before the (weighted) mean is computed. The quantiles used for trimming are defined using the wtd.quantile function.

Value

Returns the weighted and/or trimmed arithmetic mean.

Note

The weighted (and possible trimmed) mean is defined as:

sum(weights * x) / sum(weights)

where x is the (possibly trimmed version of the) input data.

Author(s)

Nathaniel E. Helwig <[e-mail]>

See Also

wtd.var for weighted variance calculations

wtd.quantile for weighted quantile calculations

Examples

# generate data and weights
set.seed(1)
x <- rnorm(10)
w <- rpois(10, lambda = 10)

# weighted mean
wtd.mean(x, w)
sum(x * w) / sum(w)

# trimmed mean
q <- quantile(x, probs = c(0.1, 0.9), type = 4)
i <- which(x < q[1] | x > q[2])
mean(x[-i])
wtd.mean(x, trim = 0.1)

# weighted and trimmed mean
q <- wtd.quantile(x, w, probs = c(0.1, 0.9))
i <- which(x < q[1] | x > q[2])
wtd.mean(x[-i], w[-i])
wtd.mean(x, w, trim = 0.1)

npreg documentation built on May 29, 2024, 4:17 a.m.