weighted_mean_winsorized: Weighted Winsorized Mean and Total (bare-bone functions)
In robsurvey: Robust Survey Statistics Estimation

View source: R/weighted_mean_winsorized.R

weighted_mean_winsorized

R Documentation

Weighted Winsorized Mean and Total (bare-bone functions)

Description

Weighted winsorized mean and total (bare-bone functions with limited functionality; see svymean_winsorized and svytotal_winsorized for more capable methods)

Usage

weighted_mean_winsorized(x, w, LB = 0.05, UB = 1 - LB, info = FALSE,
    na.rm = FALSE)
weighted_mean_k_winsorized(x, w, k, info = FALSE, na.rm = FALSE)
weighted_total_winsorized(x, w, LB = 0.05, UB = 1 - LB, info = FALSE,
    na.rm = FALSE)
weighted_total_k_winsorized(x, w, k, info = FALSE, na.rm = FALSE)

Arguments

`x`	`[numeric vector]` data.
`w`	`[numeric vector]` weights (same length as `x`).
`LB`	`[double]` lower bound of winsorization such that 0 ≤q `LB` < `UB` ≤q 1.
`UB`	`[double]` upper bound of winsorization such that 0 ≤q `LB` < `UB` ≤q 1.
`info`	`[logical]` indicating whether additional information should be returned (default: `FALSE`).
`na.rm`	`[logical]` indicating whether `NA` values should be removed before the computation proceeds (default: `FALSE`).
`k`	`[integer]` number of observations to be winsorized at the top of the distribution.

Details

Characteristic.

Population mean or total. Let μ denote the estimated winsorized population mean; then, the estimated population total is given by Nhat μ with Nhat = sum(w[i]), where summation is over all observations in the sample.

Modes of winsorization.

The amount of winsorization can be specified in relative or absolute terms:

Relative: By specifying LB and UB, the methods winsorizes the LB~\cdot 100\% of the smallest observations and the (1 - UB)~\cdot 100\% of the largest observations from the data.
Absolute: By specifying argument k in the functions with the "infix" _k_ in their name, the largest k observations are winsorized, 0<k<n, where n denotes the sample size. E.g., k = 2 implies that the largest and the second largest observation are winsorized.

Variance estimation.

See survey methods:

svymean_winsorized,
svytotal_winsorized,
svymean_k_winsorized,
svytotal_k_winsorized.

Value

The return value depends on info:

info = FALSE:

estimate of mean or total [double]

info = TRUE:

a [list] with items:

characteristic [character],
estimator [character],
estimate [double],
variance (default: NA),
robust [list],
residuals [numeric vector],
model [list],
design (default: NA),
[call]

Examples

data(workplace)

# Estimated winsorized population mean (5% symmetric winsorization)
weighted_mean_winsorized(workplace$employment, workplace$weight, LB = 0.05)

# Estimated one-sided k winsorized population total (2 observations are
# winsorized at the top of the distribution)
weighted_total_k_winsorized(workplace$employment, workplace$weight, k = 2)

robsurvey documentation built on Jan. 6, 2023, 5:09 p.m.