tests/testthat/test-gls.R

In strengejacke/ggeffects: Create Tidy Data Frames of Marginal Effects for 'ggplot' from Model Outputs

skip_on_os(c("mac", "solaris"))
skip_if_not_installed("nlme")
skip_if_not_installed("effects")
skip_if_not_installed("emmeans")

test_that("ggpredict", {
  data(Ovary, package = "nlme")
  m1 <- nlme::gls(follicles ~ Time, Ovary, correlation = nlme::corAR1(form = ~ 1 | Mare))
  p <- ggpredict(m1, "Time")
  expect_equal(p$predicted[1], 11.49246, tolerance = 1e-3)
  p <- ggeffect(m1, "Time")
  expect_equal(p$predicted[1], 11.49246, tolerance = 1e-3)
  p <- ggemmeans(m1, "Time")
  expect_equal(p$predicted[1], 11.49246, tolerance = 1e-3)
})