heteroTests v0.15.0

What's Changed

• Give every test the (model, data, ...) convention by @DiogoRibeiro7 in #115
• Call the studentizing argument studentize everywhere by @DiogoRibeiro7 in #116
• Rename rrunAdvancedDiagnostics() to runAdvancedDiagnostics() by @DiogoRibeiro7 in #117
• Document the fields every test returns and check every test against them by @DiogoRibeiro7 in #118
• Meet the same bad input with the same error in every test by @DiogoRibeiro7 in #119
• Let every test be called without data, and say the old forms go in 1.0.0 by @DiogoRibeiro7 in #120
• Draw 999 replicates by default in every resampling test by @DiogoRibeiro7 in #121
• Make White and Koenker the default battery by @DiogoRibeiro7 in #122
• Refuse a glm that is not a least-squares fit, an NA coefficient and unmatched data by @DiogoRibeiro7 in #123
• Make seeded parallel resampling reproducible by @DiogoRibeiro7 in #124
• Correct what the 0.15.0 documentation promises about optional data, the runner and the wild bootstrap by @DiogoRibeiro7 in #125
• Measure the wild bootstrap with Rademacher and with Mammen multipliers by @DiogoRibeiro7 in #126
• Refuse a non-Gaussian glm through the deprecated registry names by @DiogoRibeiro7 in #128
• Make robust = TRUE the default of performGlejserTest() by @DiogoRibeiro7 in #127
• Refuse data whose rows do not match the fitted model by @DiogoRibeiro7 in #130
• Estimate the weights of a fitWLS() fit again in each bootstrap replicate by @DiogoRibeiro7 in #129
• Release 0.15.0 by @DiogoRibeiro7 in #131

Full Changelog: v0.14.0...v0.15.0

heteroTests v0.14.0

What's Changed

• Record the 0.14.0 decisions and the defects left for 0.16.0 by @DiogoRibeiro7 in #103
• Add the export inventory to the 0.14.0 section of the roadmap by @DiogoRibeiro7 in #104
• Deprecate performBreuschPaganTest() in favour of performBPTest() by @DiogoRibeiro7 in #105
• Deprecate performStudentizedBPTest() in favour of performKoenkerTest() by @DiogoRibeiro7 in #106
• Make performCookWeisbergTest() the one entry point and deprecate performNCVTest() by @DiogoRibeiro7 in #113
• Fold the *Enhanced plots and model check into their base names by @DiogoRibeiro7 in #107
• Deprecate the exported internal helpers by @DiogoRibeiro7 in #108
• Rename rbootstrap_test_statistic() and deprecate the r-prefixed helpers and test() by @DiogoRibeiro7 in #110
• Rename the simulation one-offs into the snake_case family by @DiogoRibeiro7 in #111
• Group the reference index, with the model checks apart from the tests by @DiogoRibeiro7 in #109
• Assert the planned 1.0 export list in test-public-api.R by @DiogoRibeiro7 in #112
• Release 0.14.0 by @DiogoRibeiro7 in #114

Full Changelog: v0.13.0...v0.14.0

heteroTests v0.13.0

What's Changed

• paper: fix PDF formatting, figures and render checks by @DiogoRibeiro7 in #76
• paper: update the manuscript for 0.12.0 by @DiogoRibeiro7 in #77
• Fix grouped refits and the handling of survey designs by @DiogoRibeiro7 in #78
• paper: address a reader's comments on the manuscript by @DiogoRibeiro7 in #79
• Add the tests found in the literature to the roadmap by @DiogoRibeiro7 in #80
• Fix autoplot() and tidy() for test suites by @DiogoRibeiro7 in #81
• paper: use the package's plotting functions for two figures by @DiogoRibeiro7 in #82
• Add the Glejser test that is valid under skewed errors by @DiogoRibeiro7 in #83
• paper: report the Glejser test for skewed errors by @DiogoRibeiro7 in #84
• Remove the definition of suggestDiagnosticsForProfile() that never ran by @DiogoRibeiro7 in #85
• Add listDiagnostics(), the names runHeteroTests() accepts by @DiogoRibeiro7 in #86
• Bring the README, citation metadata and one help page in line with the package by @DiogoRibeiro7 in #87
• Run R CMD check --as-cran on the release tarball before tagging by @DiogoRibeiro7 in #88
• Order the registered Szroeter test by the fitted values by @DiogoRibeiro7 in #89
• Record why the modified Wald test was not adopted by @DiogoRibeiro7 in #90
• Add the Harrison-McCabe test with its exact p-value by @DiogoRibeiro7 in #91
• Let the resampling tests accept a weighted fit by @DiogoRibeiro7 in #92
• paper: regenerate the committed TeX from the current article by @DiogoRibeiro7 in #93
• paper: bring the article in line with the package after 0.12.0 by @DiogoRibeiro7 in #94
• Move the R Journal article to its own repository by @DiogoRibeiro7 in #95
• Run the parallel bootstrap example only within two processes by @DiogoRibeiro7 in #96
• Keep at most 1000 results in the validation cache by @DiogoRibeiro7 in #97
• Stage 1 of the Dette-Neumeyer-Van Keilegom variance-function test by @DiogoRibeiro7 in #98
• Keep the offset when fitWLS() refits a model by @DiogoRibeiro7 in #99
• Keep the offset and the rows of the data when the resampling tests refit by @DiogoRibeiro7 in #101
• Regenerate a transformed response in the White bootstrap by @DiogoRibeiro7 in #102
• Release 0.13.0 by @DiogoRibeiro7 in #100

Full Changelog: v0.12.0...v0.13.0

heteroTests v0.12.0

What's Changed

• docs: roadmap from 0.11.2 to 1.0.0 by @DiogoRibeiro7 in #40
• Document CRAN installation and canonical package links by @DiogoRibeiro7 in #39
• ci: reuse shared R setup for pkgdown by @DiogoRibeiro7 in #41
• ci: reuse shared R setup for minimum version check by @DiogoRibeiro7 in #42
• ci: reuse shared R setup for CRAN checks by @DiogoRibeiro7 in #43
• ci: reuse shared R setup for revdep checks by @DiogoRibeiro7 in #44
• ci: reuse shared R setup for project quality by @DiogoRibeiro7 in #45
• ci: reuse shared R setup for docs quality by @DiogoRibeiro7 in #46
• ci: reuse shared R setup for benchmark checks by @DiogoRibeiro7 in #47
• ci: reuse shared R setup for releases by @DiogoRibeiro7 in #48
• paper: migrate R Journal manuscript to rjtools by @DiogoRibeiro7 in #50
• paper: write introduction and package design by @DiogoRibeiro7 in #52
• paper: write validation strategy by @DiogoRibeiro7 in #54
• paper: build reproducible simulation study by @DiogoRibeiro7 in #56
• paper: run and publish simulation results by @DiogoRibeiro7 in #58
• paper: build reproducible performance study by @DiogoRibeiro7 in #60
• paper: integrate empirical simulation and performance results by @DiogoRibeiro7 in #62
• paper: add reproducible diagnostic workflow by @DiogoRibeiro7 in #64
• paper: write comparison with existing R software by @DiogoRibeiro7 in #66
• ci: remove R Journal study workflows by @DiogoRibeiro7 in #67
• paper: write discussion and conclusion by @DiogoRibeiro7 in #69
• paper: prepare R Journal submission tree by @DiogoRibeiro7 in #71
• paper: fix WLS figure caption LaTeX error by @DiogoRibeiro7 in #72
• paper: work around rjtools spelling parser failure by @DiogoRibeiro7 in #73
• paper: run R Journal checks individually by @DiogoRibeiro7 in #74
• Compute tests on weighted fits from the Pearson residuals and add performVarianceFormTest() by @DiogoRibeiro7 in #75

Full Changelog: v0.11.2...v0.12.0

heteroTests v0.11.2

What's Changed

• Add a heavy-tailed size column to the sweep by @DiogoRibeiro7 in #35
• Store CRAN source tarball on every release by @DiogoRibeiro7 in #36
• Address the CRAN review of the 0.11.1 submission by @DiogoRibeiro7 in #37

Full Changelog: v0.11.1...v0.11.2

heteroTests v0.11.1

What's Changed

• Publish the full-sweep size and power table by @DiogoRibeiro7 in #31
• Check the panel statistics against plm, and stop their p-values underflowing by @DiogoRibeiro7 in #32
• Stop running the CRAN check on every merge by @DiogoRibeiro7 in #33
• Record the changes that landed after v0.11.0 was released by @DiogoRibeiro7 in #34

Full Changelog: v0.10.0...v0.11.1

heteroTests v0.11.0

What's Changed

• Resample under the null in rbootstrap_test_statistic by @DiogoRibeiro7 in #29
• Correct two panel statistics against their published definitions by @DiogoRibeiro7 in #30

Full Changelog: v0.9.0...v0.11.0

heteroTests v0.10.0

What's Changed

• Resample under the null in rbootstrap_test_statistic by @DiogoRibeiro7 in #29
• Correct two panel statistics against their published definitions by @DiogoRibeiro7 in #30

Full Changelog: v0.9.0...v0.10.0

heteroTests v0.9.0

What's Changed

• Estimate fitWLS() weights from a variance model by @DiogoRibeiro7 in #28

Full Changelog: v0.8.1...v0.9.0

heteroTests v0.8.1

What's Changed

• Replace the bundled Boston dataset with R's built-in quakes by @DiogoRibeiro7 in #27

Full Changelog: v0.8.0...v0.8.1