tests/testthat/test-rss.sampling.R

In generalRSS: Statistical Tools for Balanced and Unbalanced Ranked Set Sampling

library(testthat)

test_that("rss.sampling works as expected",{
  data(iris)
  id=1:nrow(iris)
  result <- rss.sampling(ID=id, Y=iris$Sepal.Length, X=iris$Petal.Length, H=3, nsamp=c(3,5,7))
  expect_s3_class(result,"data.frame")
  expect_equal(colnames(result), c("rank","ID","y"))
  expect_equal(dim(result), c(15,3))
  result.lth <- as.numeric(tapply(result$y,result$rank,length))
  expect_equal(result.lth, c(3,5,7))
})

test_that("rss.sampling handles wrong input with error",{
  data(iris)
  id=1:nrow(iris)
  expect_error(rss.sampling(ID=id, Y=iris$Sepal.Length, X=iris$Petal.Length, H=5, nsamp=c(3,5,7)),
               "Set size are different with the length of sample allocations")
  expect_error(rss.sampling(ID=id, Y=iris$Sepal.Length, X=iris$Petal.Length, H=2, nsamp=c(3,5,7)),
               "Set size are different with the length of sample allocations")
  # Mismatched lengths of ID, X, Y
  expect_error(rss.sampling(ID = id[-1], Y = iris$Sepal.Length, X = iris$Petal.Length, H = 3, nsamp = c(3, 5, 7)),
               "ID and X must have the same length.")
  expect_error(rss.sampling(ID = id, Y = iris$Sepal.Length[-1], X = iris$Petal.Length, H = 3, nsamp = c(3, 5, 7)),
               "ID and Y must have the same length if Y is provided.")

  # Non-unique IDs
  expect_error(rss.sampling(ID = c(id, id), Y = c(iris$Sepal.Length, iris$Sepal.Length), X = c(iris$Petal.Length, iris$Petal.Length), H = 3, nsamp = c(3, 5, 7)),
               "ID must be unique.")

})