Release: 1.7.2

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: powerly
 Title: Sample Size Analysis for Psychological Networks and More
-Version: 1.7.1
+Version: 1.7.2
 Authors@R:
     person(given = "Mihai",
            family = "Constantin",

NEWS.md

@@ -1,3 +1,27 @@
+# `powerly` `1.7.2`
+
+## Improvements
+
+* Update GGM estimation test to check if the estimation fails when variables
+  with zero standard deviation are present in the generated data.
+
+* Update GGM estimation to fail when the generated data contains at least one
+  variable that has a standard deviation equal to zero (i.e., as a result of
+  generating data with a sample size value that is too low).
+
+* Add GitHub badges with latest release version and number of open issues.
+
+## Bug fixes
+
+* Add tolerance (i.e., `0.0000001` for test checking whether the spline
+  coefficients are estimated correctly).
+
+* Fix test for the updating of the bounds of a `Range` instance to run only when
+  the the 2.5th and 97.5th selected sample sizes are different quantities.
+
+* Fix bug in GGM data generating test where the number of nodes to generate data
+  for were incorrectly sampled.
+
 # `powerly` `1.7.1`
 
 ## Improvements

R/GgmModel.R

@@ -24,6 +24,11 @@ GgmModel <- R6::R6Class("GgmModel",
         },
 
         estimate = function(data, gamma = 0.5) {
+            # Ensure all variables show variance.
+            if (sum(apply(data, 2, sd) == 0) > 0) {
+                stop("Variable(s) with SD = 0 detected. Increase the sample size.")
+            }
+
             # Estimate network using `qgraph`.
             network <- suppressMessages(suppressWarnings(
                 qgraph::EBICglasso(

README.md

@@ -9,12 +9,14 @@
 </p>
 
 <p align="center">
-    <a align="center" href="https://github.com/mihaiconstantin/powerly/actions"><img src="https://github.com/mihaiconstantin/powerly/workflows/R-CMD-check/badge.svg" alt="R-CMD-check" /></a>
+    <a href="https://github.com/mihaiconstantin/powerly/releases"><img src="https://img.shields.io/github/v/release/mihaiconstantin/powerly?display_name=tag&sort=semver"/></a>
     <a href="https://www.r-pkg.org/pkg/powerly"><img src="https://www.r-pkg.org/badges/version/powerly" alt="CRAN version"/></a>
-    <a href="https://cranchecks.info/pkgs/powerly"><img src="https://cranchecks.info/badges/worst/powerly" alt="CRAN checks"/></a>
-    <a href="https://app.codecov.io/gh/mihaiconstantin/powerly"><img src="https://codecov.io/gh/mihaiconstantin/powerly/branch/main/graph/badge.svg?token=YUCO8ULBCM" alt="Code coverage"/></a>
     <a href="https://www.repostatus.org/#active"><img src="https://www.repostatus.org/badges/latest/active.svg" alt="Repository status"/></a>
+    <a href="https://app.codecov.io/gh/mihaiconstantin/powerly"><img src="https://codecov.io/gh/mihaiconstantin/powerly/branch/main/graph/badge.svg?token=YUCO8ULBCM" alt="Code coverage"/></a>
+    <a href="https://github.com/mihaiconstantin/powerly/issues"><img src="https://img.shields.io/github/issues/mihaiconstantin/powerly"/></a>
     <a href="https://www.r-pkg.org/pkg/powerly"><img src="https://cranlogs.r-pkg.org/badges/grand-total/powerly" alt="CRAN RStudio mirror downloads"/></a>
+    <a href="https://github.com/mihaiconstantin/powerly/actions"><img src="https://github.com/mihaiconstantin/powerly/workflows/R-CMD-check/badge.svg" alt="R-CMD-check" /></a>
+    <a href="https://cranchecks.info/pkgs/powerly"><img src="https://cranchecks.info/badges/worst/powerly" alt="CRAN checks"/></a>
 </p>
 
 ## Description

tests/testthat/test-ggm.R

@@ -8,7 +8,7 @@ test_that("'GgmModel' generates data correctly", {
     max_level <- sample(3:5, 1)
 
     # Nodes.
-    nodes <- sample(10:20)
+    nodes <- sample(10:20, 1)
 
     # Density.
     density <- sample(seq(.2, .5, .1), 1)
@@ -40,7 +40,7 @@ test_that("'GgmModel' generated data matches bootnet data", {
     max_level <- sample(3:5, 1)
 
     # Nodes.
-    nodes <- sample(10:20)
+    nodes <- sample(10:20, 1)
 
     # Density.
     density <- sample(seq(.2, .5, .1), 1)
@@ -75,7 +75,7 @@ test_that("'GgmModel' estimates model parameters correctly", {
     max_level <- sample(3:5, 1)
 
     # Nodes.
-    nodes <- sample(10:20)
+    nodes <- sample(10:20, 1)
 
     # Density.
     density <- sample(seq(.2, .5, .1), 1)
@@ -97,6 +97,12 @@ test_that("'GgmModel' estimates model parameters correctly", {
 
     # The parameters should be identical across both methods.
     expect_equal(network_qgraph, network_ggm_model, ignore_attr = TRUE)
+
+    # Make one variable invariant.
+    data[, 1] <- data[1, 1]
+
+    # Expect the estimation to throw an error due to invariant variables.
+    expect_error(ggm$estimate(data), "Variable\\(s\\) with SD = 0 detected. Increase the sample size.")
 })
 
 
@@ -108,7 +114,7 @@ test_that("'GgmModel' computes the correct measure", {
     max_level <- sample(3:5, 1)
 
     # Nodes.
-    nodes <- sample(10:20)
+    nodes <- sample(10:20, 1)
 
     # Density.
     density <- sample(seq(.2, .5, .1), 1)

tests/testthat/test-range.R

@@ -72,11 +72,14 @@ test_that("'Range' updates bounds correctly based on 'StepThree' confidence inte
     expect_equal(min(range$partition), as.numeric(step_3$samples["2.5%"]))
     expect_equal(max(range$partition), as.numeric(step_3$samples["97.5%"]))
 
-    # Expect that the bounds are of increasing size.
-    expect_error(
-        range$update_bounds(step_3, lower_ci = 0.975, upper_ci = 0.025),
-        "The lower bound cannot be greater that the upper bound."
-    )
+    # If the selected samples are not a single quantity.
+    if(step_3$samples["2.5%"] < step_3$samples["97.5%"]) {
+        # Expect that the bounds are of increasing size.
+        expect_error(
+            range$update_bounds(step_3, lower_ci = 0.975, upper_ci = 0.025),
+            "The lower bound cannot be greater that the upper bound."
+        )
+    }
 })
 
 

tests/testthat/test-step-two.R

@@ -126,7 +126,7 @@ test_that("'StepTwo' fits and interpolates a spline correctly", {
     expect_equal(step_2$spline$basis$matrix, basis)
 
     # The estimated spline coefficients should be equal.
-    expect_equal(step_2$spline$alpha, alpha)
+    expect_equal(step_2$spline$alpha, alpha, tolerance = 0.0000001)
 
     # The fitted values should be equal.
     expect_equal(step_2$spline$fitted, fitted)