safer tests because it fails on some distribution

Author: DominiqueMakowski

tests/testthat/test-cor_test.R

@@ -1,22 +1,30 @@
 context("cor_test")
 
 
-
 test_that("cor_test frequentist", {
-  if (requireNamespace("psych")) {
-    data <- iris
-    data$Sepal.Width_binary <- ifelse(data$Sepal.Width > 3, 1, 0)
-    data$Petal.Width_binary <- ifelse(data$Petal.Width > 1.2, 1, 0)
 
+  testthat::expect_error(cor_test(iris, Petal.Length, Petal.Width))
 
-    testthat::expect_error(cor_test(data, Petal.Length, Petal.Width))
+  out <- cor_test(iris, "Petal.Length", "Petal.Width")
+  testthat::expect_equal(out$r, 0.962, tol = 0.01)
+
+})
 
-    out <- cor_test(data, "Petal.Length", "Petal.Width")
-    testthat::expect_equal(out$r, 0.962, tol = 0.01)
 
-    out <- cor_test(data, "Petal.Length", "Petal.Width", bayesian = TRUE)
-    testthat::expect_equal(out$r, 0.962, tol = 0.01)
+test_that("cor_test bayesian", {
+  if (requireNamespace("BayesFactor")) {
 
+  out <- cor_test(iris, "Petal.Length", "Petal.Width", bayesian = TRUE)
+  testthat::expect_equal(out$r, 0.962, tol = 0.01)
+
+  }
+})
+
+test_that("cor_test tetrachoric", {
+  if (requireNamespace("psych")) {
+    data <- iris
+    data$Sepal.Width_binary <- ifelse(data$Sepal.Width > 3, 1, 0)
+    data$Petal.Width_binary <- ifelse(data$Petal.Width > 1.2, 1, 0)
 
     # With Factors / Binary
     out <- cor_test(data, "Sepal.Width_binary", "Petal.Width_binary", method = "tetrachoric")
@@ -85,9 +93,11 @@ test_that("cor_test blomqvist", {
 })
 
 test_that("cor_test hoeffding", {
+  if (requireNamespace("Hmisc")) {
   set.seed(333)
   out <- cor_test(iris, "Petal.Length", "Petal.Width", method = "hoeffding")
   testthat::expect_equal(out$r, as.numeric(0.5629277), tol = 0.01)
+  }
 })
 
 test_that("cor_test gamma", {
@@ -101,8 +111,10 @@ test_that("cor_test gaussian", {
   out <- cor_test(iris, "Petal.Length", "Petal.Width", method = "gaussian")
   testthat::expect_equal(out$r, as.numeric(0.87137), tol = 0.01)
 
-  out <- cor_test(iris, "Petal.Length", "Petal.Width", method = "gaussian", bayesian = TRUE)
-  testthat::expect_equal(out$r, as.numeric(0.8620878), tol = 0.01)
+  if (requireNamespace("BayesFactor")) {
+    out <- cor_test(iris, "Petal.Length", "Petal.Width", method = "gaussian", bayesian = TRUE)
+    testthat::expect_equal(out$r, as.numeric(0.8620878), tol = 0.01)
+  }
 })
 
 

tests/testthat/test-correlation.R

@@ -68,40 +68,42 @@ test_that("comparison with other packages", {
 
 
     # Bayesian
-    out <- correlation(iris, include_factors = FALSE, bayesian = TRUE)
-    rez <- as.data.frame(as.table(out))
+    if (requireNamespace("BayesFactor")) {
+      out <- correlation(iris, include_factors = FALSE, bayesian = TRUE)
+      rez <- as.data.frame(as.table(out))
 
-    r <- as.matrix(rez[2:5])
-    testthat::expect_equal(mean(r - cor(iris[1:4])), 0, tol = 0.01)
+      r <- as.matrix(rez[2:5])
+      testthat::expect_equal(mean(r - cor(iris[1:4])), 0, tol = 0.01)
 
-    hmisc <- Hmisc::rcorr(as.matrix(iris[1:4]), type = c("pearson"))
-    testthat::expect_equal(mean(r - hmisc$r), 0, tol = 0.01)
+      hmisc <- Hmisc::rcorr(as.matrix(iris[1:4]), type = c("pearson"))
+      testthat::expect_equal(mean(r - hmisc$r), 0, tol = 0.01)
 
-    pd <- as.matrix(attributes(rez)$pd[2:5])
-    p <- bayestestR::pd_to_p(pd)
-    testthat::expect_equal(mean(p - hmisc$P, na.rm = TRUE), 0, tol = 0.01)
+      pd <- as.matrix(attributes(rez)$pd[2:5])
+      p <- bayestestR::pd_to_p(pd)
+      testthat::expect_equal(mean(p - hmisc$P, na.rm = TRUE), 0, tol = 0.01)
 
 
-    # Bayesian - Partial
-    out <- correlation(iris, include_factors = FALSE, bayesian = TRUE, partial = TRUE)
-    rez <- as.data.frame(as.table(out))
+      # Bayesian - Partial
+      out <- correlation(iris, include_factors = FALSE, bayesian = TRUE, partial = TRUE)
+      rez <- as.data.frame(as.table(out))
 
-    r <- as.matrix(rez[2:5])
-    ppcor <- ppcor::pcor(iris[1:4])
-    testthat::expect_equal(max(r - as.matrix(ppcor$estimate)), 0, tol = 0.02)
+      r <- as.matrix(rez[2:5])
+      ppcor <- ppcor::pcor(iris[1:4])
+      testthat::expect_equal(max(r - as.matrix(ppcor$estimate)), 0, tol = 0.02)
 
-    pd <- as.matrix(attributes(rez)$pd[2:ncol(rez)])
-    p <- bayestestR::pd_to_p(pd)
-    testthat::expect_equal(mean(abs(p - as.matrix(ppcor$p.value))), 0, tol = 0.001)
+      pd <- as.matrix(attributes(rez)$pd[2:ncol(rez)])
+      p <- bayestestR::pd_to_p(pd)
+      testthat::expect_equal(mean(abs(p - as.matrix(ppcor$p.value))), 0, tol = 0.001)
 
 
-    # Bayesian (Full) - Partial
-    out <- correlation(iris, include_factors = FALSE, bayesian = TRUE, partial = TRUE, partial_bayesian = TRUE)
-    rez <- as.data.frame(as.table(out))
+      # Bayesian (Full) - Partial
+      out <- correlation(iris, include_factors = FALSE, bayesian = TRUE, partial = TRUE, partial_bayesian = TRUE)
+      rez <- as.data.frame(as.table(out))
 
-    r <- as.matrix(rez[2:5])
-    ppcor <- ppcor::pcor(iris[1:4])
-    testthat::expect_equal(max(r - as.matrix(ppcor$estimate)), 0, tol = 0.02)
+      r <- as.matrix(rez[2:5])
+      ppcor <- ppcor::pcor(iris[1:4])
+      testthat::expect_equal(max(r - as.matrix(ppcor$estimate)), 0, tol = 0.02)
+    }
   }
 })
 
@@ -140,10 +142,12 @@ test_that("format checks", {
   }
 
   # Bayesian full partial
-  out <- correlation(iris, include_factors = TRUE, multilevel = TRUE, bayesian = TRUE, partial = TRUE, partial_bayesian = TRUE)
-  testthat::expect_equal(c(nrow(out), ncol(out)), c(6, 13))
-  testthat::expect_equal(c(nrow(as.table(out)), ncol(as.table(out))), c(4, 5))
-  testthat::expect_equal(c(nrow(summary(out)), ncol(summary(out))), c(3, 4))
+  if (requireNamespace("BayesFactor")) {
+    out <- correlation(iris, include_factors = TRUE, multilevel = TRUE, bayesian = TRUE, partial = TRUE, partial_bayesian = TRUE)
+    testthat::expect_equal(c(nrow(out), ncol(out)), c(6, 13))
+    testthat::expect_equal(c(nrow(as.table(out)), ncol(as.table(out))), c(4, 5))
+    testthat::expect_equal(c(nrow(summary(out)), ncol(summary(out))), c(3, 4))
+  }
 })