p_function() gets vcov-args

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: parameters
 Title: Processing of Model Parameters
-Version: 0.22.2.7
+Version: 0.22.2.8
 Authors@R:
     c(person(given = "Daniel",
              family = "Lüdecke",

NEWS.md

@@ -9,6 +9,11 @@
   way to test the direction of the effect, which formerly was already (and still
   is) possible with `pd = TRUE` in `model_parameters()`.
 
+* `p_function()` gets a `vcov` and `vcov_args` argument to compute robust
+  standard errors for the confidence curves.
+
+* Revision / enhancement of some documentation.
+
 # parameters 0.22.2
 
 ## New supported models

R/p_function.R

@@ -18,6 +18,7 @@
 #' @inheritParams model_parameters
 #' @inheritParams model_parameters.default
 #' @inheritParams model_parameters.glmmTMB
+#' @inheritParams standard_error
 #'
 #' @note
 #' Curently, `p_function()` computes intervals based on Wald t- or z-statistic.
@@ -61,7 +62,7 @@
 #' (i.e. assumptions are taken as given). The unconditional interpretation (B),
 #' however, questions all these assumptions.
 #'
 #' \if{html}{\cr \figure{unconditional_interpretation.png}{options: alt="Conditional versus unconditional interpretations of P-values"} \cr}
 #'
 #' "Emphasizing unconditional interpretations helps avoid overconfident and
 #' misleading inferences in light of uncertainties about the assumptions used
@@ -223,6 +224,8 @@
                        exponentiate = FALSE,
                        effects = "fixed",
                        component = "all",
+                       vcov = NULL,
+                       vcov_args = NULL,
                        keep = NULL,
                        drop = NULL,
                        verbose = TRUE,
@@ -234,7 +237,9 @@
   se <- standard_error(
     model,
     effects = effects,
-    component = component
+    component = component,
+    vcov = vcov,
+    vcov_args = vcov_args
   )$SE
 
   if (is.null(dof) || length(dof) == 0 || .is_chi2_model(model, dof)) {

tests/testthat/test-p_function.R

@@ -1,8 +1,18 @@
+skip_if_not_installed("sandwich")
+
 data(iris)
 model <- lm(Sepal.Length ~ Species, data = iris)
 
 test_that("p_function ci-levels", {
   out <- p_function(model)
+  expect_equal(
+    out$CI_low,
+    c(
+      4.982759, 0.897132, 1.549132, 4.956774, 0.860384, 1.512384,
+      4.92192, 0.811093, 1.463093, 4.862126, 0.726531, 1.378531
+    ),
+    tolerance = 1e-4
+  )
 
   expect_identical(dim(out), c(12L, 5L))
 
@@ -32,6 +42,16 @@ test_that("p_function ci-levels", {
     c(0.3, 0.3, 0.3, 0.6, 0.6, 0.6, 0.9, 0.9, 0.9),
     tolerance = 1e-4
   )
+
+  out <- p_function(model, vcov = "HC3")
+  expect_equal(
+    out$CI_low,
+    c(
+      4.989925, 0.901495, 1.548843, 4.971951, 0.869624, 1.511772,
+      4.947844, 0.826875, 1.462047, 4.906485, 0.753538, 1.376742
+    ),
+    tolerance = 1e-4
+  )
 })