rss.prop.sampling {generalRSS} | R Documentation |
Generate ranked set samples for proportions
Description
The rss.prop.sampling function generates ranked set samples for proportions by performing ranked set sampling directly on a given population data set using an auxiliary variable (X) and subject IDs.
Usage
rss.prop.sampling(ID, X, H, nsamp)
Arguments
ID |
A numeric vector of subject IDs from the population. IDs must be unique. |
X |
A numeric vector of auxiliary variable used for ranking. Must have the same length as ID. |
H |
The RSS set size |
nsamp |
A numeric vector specifying the sample allocation for each stratum. |
Details
This function performs balanced or unbalanced ranked set sampling for proportions from a given data set. The length of the sample allocation vector (nsamp) must match the set size (H). The subject ID and auxiliary variable (X) must have the same length.
Value
A data frame with the following columns:
ID |
The sampled subjects' IDs. |
rank |
The rank information assigned to each sample. |
Examples
## Example 1: Balanced RSS with equal sample sizes.
data(iris)
id=1:nrow(iris)
X=ifelse(iris$Sepal.Length<5.8,0,1)
rss.prop.data=rss.prop.sampling(ID=id, X=X, H=3,nsamp=c(6,6,6))
## Example 2: Unbalanced RSS with different sample sizes.
rss.prop.data=rss.prop.sampling(ID=id, X=X, H=3, nsamp=c(6,10,8))
# Check the structure of the RSS data
colnames(rss.prop.data) # include "ID", "rank", and "Y"
head(rss.prop.data$ID)
head(rss.prop.data$rank)