Prepare CRAN submission

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: insight
 Title: Easy Access to Model Information for Various Model Objects
-Version: 1.4.2.13
+Version: 1.4.3
 Authors@R:
     c(person(given = "Daniel",
              family = "Lüdecke",

NEWS.md

@@ -1,4 +1,4 @@
-# insight (devel)
+# insight 1.4.3
 
 ## Changes
 

R/is_converged.R

@@ -4,7 +4,8 @@
 #' @description `is_converged()` provides an alternative convergence
 #'   test for `merMod`-objects.
 #'
-#' @param x A model object from class `merMod`, `glmmTMB`, `glm` or `_glm`.
+#' @param x A model object from class `merMod`, `glmmTMB`, `glm`, `lavaan` or
+#' `_glm`.
 #' @param tolerance Indicates up to which value the convergence result is
 #'   accepted. The smaller `tolerance` is, the stricter the test will be.
 #' @param ... Currently not used.
@@ -26,23 +27,34 @@
 #' convergence test for `merMod`-objects.
 #'
 #' @section Resolving convergence issues:
-#' Convergence issues are not easy to diagnose. The help page on `?lme4::convergence`
-#' provides most of the current advice about how to resolve convergence issues.
-#' Another clue might be large parameter values, e.g. estimates (on the scale of
-#' the linear predictor) larger than 10 in (non-identity link) generalized linear
-#' model *might* indicate complete separation, which can be addressed by
-#' regularization, e.g. penalized regression or Bayesian regression with
-#' appropriate priors on the fixed effects.
+#' Convergence issues are not easy to diagnose. The help page on
+#' `?lme4::convergence` provides most of the current advice about how to resolve
+#' convergence issues. In general, convergence issues may be addressed by one or
+#' more of the following strategies: 1. Rescale continuous predictors; 2. try a
+#' different optimizer; 3. increase the number of iterations; or, if everything
+#' else fails, 4. simplify the model. Another clue might be large parameter
+#' values, e.g. estimates (on the scale of the linear predictor) larger than 10
+#' in (non-identity link) generalized linear model *might* indicate complete
+#' separation, which can be addressed by regularization, e.g. penalized
+#' regression or Bayesian regression with appropriate priors on the fixed
+#' effects.
 #'
 #' @section Convergence versus Singularity:
 #' Note the different meaning between singularity and convergence: singularity
 #' indicates an issue with the "true" best estimate, i.e. whether the maximum
 #' likelihood estimation for the variance-covariance matrix of the random effects
 #' is positive definite or only semi-definite. Convergence is a question of
 #' whether we can assume that the numerical optimization has worked correctly
-#' or not.
+#' or not. A convergence failure means the optimizer (the algorithm) could not
+#' find a stable solution (_Bates et. al 2015_).
+#'
+#' @references
+#' Bates, D., Mächler, M., Bolker, B., and Walker, S. (2015). Fitting Linear
+#' Mixed-Effects Models Using lme4. Journal of Statistical Software, 67(1),
+#' 1-48. \doi{10.18637/jss.v067.i01}
 #'
 #' @examplesIf require("lme4", quietly = TRUE)
+#' library(lme4)
 #' data(cbpp)
 #' set.seed(1)
 #' cbpp$x <- rnorm(nrow(cbpp))
@@ -58,6 +70,7 @@
 #'
 #' @examplesIf getOption("warn") < 2L && require("glmmTMB")
 #' \donttest{
+#' library(glmmTMB)
 #' model <- glmmTMB(
 #'   Sepal.Length ~ poly(Petal.Width, 4) * poly(Petal.Length, 4) +
 #'     (1 + poly(Petal.Width, 4) | Species),
@@ -108,7 +121,7 @@
 is_converged.glm <- function(x, tolerance = 0.001, ...) {
   if (!is.null(x$converged)) {
     isTRUE(x$converged)
   } else if (!is.null(x$fit$converged)) {
     isTRUE(x$fit$converged)
   } else {
     NULL

tests/testthat/test-averaging.R

@@ -1,6 +1,6 @@
+skip_on_cran()
 skip_if_not_installed("MuMIn")
 skip_if_not_installed("withr")
-skip_on_cran()
 
 withr::with_options(
   list(na.action = "na.fail"),

tests/testthat/test-coxme.R

@@ -1,10 +1,10 @@
+skip_on_cran()
 skip_if_not_installed("survival")
 skip_if_not_installed("lme4")
 skip_if_not_installed("nlme")
 skip_if_not_installed("bdsmatrix")
 skip_if_not_installed("coxme")
 skip_if_not_installed("withr")
-skip_on_cran()
 
 withr::with_environment(
   new.env(),

tests/testthat/test-get_priors.R

@@ -1,6 +1,7 @@
+skip_on_cran()
+
 test_that("get_priors", {
   skip_on_os(os = c("mac", "windows"))
-  skip_on_cran()
   skip_if_not_installed("brms")
 
   set.seed(123)
@@ -9,13 +10,12 @@
   priors <- insight::get_priors(model)
 
   expect_equal(priors$Location, c(19.2, NA, 0), tolerance = 1e-3)
   expect_equal(priors$Distribution, c("student_t", "uniform", "student_t"))
   expect_equal(priors$Parameter, c("b_Intercept", "b_wt", "sigma"))
 })
 
 test_that("get_priors, stanmvref", {
   skip_on_os(os = "mac")
-  skip_on_cran()
   skip_if_not_installed("curl")
   skip_if_offline()
   skip_if_not_installed("httr2")

tests/testthat/test-mlogit.R

@@ -1,5 +1,5 @@
-skip_if_not_installed("mlogit")
 skip_on_cran()
+skip_if_not_installed("mlogit")
 
 data("Fishing", package = "mlogit")
 Fish <-

tests/testthat/test-phylolm.R

@@ -1,8 +1,8 @@
+skip_on_cran()
 skip_if_not_installed("phylolm")
 skip_if_not_installed("ape")
 skip_if_not_installed("MuMIn")
 skip_if_not_installed("withr")
-skip_on_cran()
 
 withr::with_options(
   list(na.action = "na.fail"),

tests/testthat/test-rms.R

@@ -1,3 +1,4 @@
+skip_on_cran()
 skip_if_not_installed("rms")
 
 data(mtcars)

tests/testthat/test-tidymodels.R

@@ -1,5 +1,7 @@
+skip_on_cran()
 skip_if_not_installed("parsnip")
 
+data(mtcars)
 m <- parsnip::linear_reg()
 m <- parsnip::set_engine(m, "lm")
 m <- parsnip::set_mode(m, "regression")
@@ -13,129 +15,63 @@
   )
 })
 
-# test_that("model_info", {
-#   expect_true(model_info(m1)$is_poisson)
-#   expect_true(model_info(m1)$is_count)
-#   expect_false(model_info(m1)$is_negbin)
-#   expect_false(model_info(m1)$is_binomial)
-#   expect_false(model_info(m1)$is_linear)
-# })
-#
-# test_that("loglik", {
-#   expect_equal(get_loglikelihood(m1), logLik(m1), ignore_attr = TRUE)
-# })
-#
-# test_that("get_df", {
-#   expect_equal(get_df(m1), df.residual(m1), ignore_attr = TRUE)
-#   expect_equal(get_df(m1, type = "model"), attr(logLik(m1), "df"), ignore_attr = TRUE)
-# })
-#
-#
-# test_that("find_predictors", {
-#   expect_identical(find_predictors(m1), list(conditional = c("mined", "cover", "sample")))
-#   expect_identical(
-#     find_predictors(m1, flatten = TRUE),
-#     c("mined", "cover", "sample")
-#   )
-#   expect_null(find_predictors(m1, effects = "random"))
-# })
-#
-# test_that("find_random", {
-#   expect_null(find_random(m1))
-# })
-#
-# test_that("get_random", {
-#   expect_warning(get_random(m1))
-# })
-#
-# test_that("find_response", {
-#   expect_identical(find_response(m1), "count")
-# })
-#
-# test_that("get_response", {
-#   expect_equal(get_response(m1), Salamanders$count)
-# })
-#
-# test_that("get_predictors", {
-#   expect_equal(colnames(get_predictors(m1)), c("mined", "cover", "sample"))
-# })
-#
-# test_that("link_inverse", {
-#   expect_equal(link_inverse(m1)(0.2), exp(0.2), tolerance = 1e-5)
-# })
-#
-# test_that("linkfun", {
-#   expect_equal(link_function(m1)(0.2), -1.609438, tolerance = 1e-4)
-# })
-#
-# test_that("get_data", {
-#   expect_equal(nrow(get_data(m1)), 644)
-#   expect_equal(
-#     colnames(get_data(m1)),
-#     c("count", "mined", "cover", "sample")
-#   )
-# })
-#
-# test_that("get_call", {
-#   expect_equal(class(get_call(m1)), "call")
-# })
-#
-#
-#
-# test_that("find_variables", {
-#   expect_equal(
-#     find_variables(m1),
-#     list(
-#       response = "count",
-#       conditional = c("mined", "cover", "sample")
-#     )
-#   )
-#   expect_equal(
-#     find_variables(m1, flatten = TRUE),
-#     c("count", "mined", "cover", "sample")
-#   )
-# })
-#
-# test_that("n_obs", {
-#   expect_equal(n_obs(m1), 644)
-# })
-#
-# test_that("find_parameters", {
-#   expect_equal(
-#     find_parameters(m1),
-#     list(
-#       conditional = c("(Intercept)", "minedno", "log(cover)", "sample")
-#     )
-#   )
-#   expect_equal(nrow(get_parameters(m1)), 4)
-#   expect_equal(
-#     get_parameters(m1)$Parameter,
-#     c("(Intercept)", "minedno", "log(cover)", "sample")
-#   )
-# })
-#
-# test_that("is_multivariate", {
-#   expect_false(is_multivariate(m1))
-# })
-#
-# test_that("find_terms", {
-#   expect_equal(
-#     find_terms(m1),
-#     list(
-#       response = "count",
-#       conditional = c("mined", "log(cover)", "sample")
-#     )
-#   )
-# })
-#
-# test_that("find_algorithm", {
-#   expect_equal(find_algorithm(m1), list(algorithm = "ML"))
-# })
-#
-# test_that("find_statistic", {
-#   expect_identical(find_statistic(m1), "z-statistic")
-# })
-#
-# test_that("get_statistic", {
-#   expect_equal(get_statistic(m1)$Statistic, c(-10.7066515607315, 18.1533878215937, -1.68918157150882, 2.23541768590273), tolerance = 1e-4)
-# })
+test_that("model_info", {
+  expect_true(model_info(m)$is_linear)
+})
+
+test_that("loglik", {
+  expect_equal(
+    get_loglikelihood(m),
+    -83.8397585518224,
+    tolerance = 1e-4,
+    ignore_attr = TRUE
+  )
+})
+
+test_that("get_df", {
+  expect_equal(get_df(m), 29, ignore_attr = TRUE)
+  expect_equal(get_df(m, type = "model"), 4, ignore_attr = TRUE)
+})
+
+
+test_that("find_predictors", {
+  expect_identical(find_predictors(m), list(conditional = c("am", "vs")))
+  expect_identical(
+    find_predictors(m, flatten = TRUE),
+    c("am", "vs")
+  )
+  expect_null(find_predictors(m, effects = "random"))
+})
+
+test_that("find_random", {
+  expect_null(find_random(m))
+})
+
+test_that("get_random", {
+  expect_warning(get_random(m))
+})
+
+test_that("find_response", {
+  expect_identical(find_response(m), "mpg")
+})
+
+test_that("get_response", {
+  expect_equal(get_response(m), mtcars$mpg)
+})
+
+test_that("get_predictors", {
+  expect_equal(colnames(get_predictors(m)), c("am", "vs"))
+})
+
+test_that("link_inverse", {
+  expect_equal(link_inverse(m)(0.2), 0.2, tolerance = 1e-5)
+})
+
+test_that("linkfun", {
+  expect_equal(link_function(m)(0.2), -1.609438, tolerance = 1e-4)
+})
+
+test_that("get_data", {
+  expect_equal(nrow(get_data(m)), 32)
+  expect_named(get_data(m), c("mpg", "am", "vs"))
+})