m <- lm(formula = wt ~ am * cyl * vs, data = mtcars)

test_that("n_parameters-rank_deficiency", {
  expect_equal(n_parameters(m), 8)
  expect_equal(n_parameters(m, remove_nonestimable = TRUE), m$rank)
})