aboutsummaryrefslogtreecommitdiff
path: root/tests/testthat
AgeCommit message (Collapse)AuthorFilesLines
2019-02-25Add plotting tests, add nafta to pkgdownJohannes Ranke1-7/+19
2019-02-22Implement the NAFTA evaluation schemeJohannes Ranke1-0/+41
2019-02-22Adapt test to the extended endpoints functionJohannes Ranke1-2/+2
2019-02-21Remove unused codeJohannes Ranke1-1/+0
2019-02-21Add the logistic modelJohannes Ranke1-0/+47
2019-01-31Skipping with skip() needs a messageJohannes Ranke1-1/+1
2019-01-31Skip tests for irls methd tc because it is not robust enoughJohannes Ranke1-1/+3
2019-01-31Relax yet another testJohannes Ranke1-1/+1
2019-01-31Relax another test for TravisJohannes Ranke1-1/+1
2018-11-26See last commit logJohannes Ranke1-3/+3
2018-11-26Another test relaxationJohannes Ranke1-1/+1
This feels stupid by now, but I am not sure how to do this in a smarter way
2018-11-26Maybe I should not used random data in tests...Johannes Ranke1-1/+1
2018-11-26Relax testing strictness again for travisJohannes Ranke1-1/+1
2018-11-23Relax yet another test...Johannes Ranke1-1/+1
2018-11-23More tolerance for a testJohannes Ranke1-1/+1
2018-11-23Add logLik method to enable AIC() on mkinfit modelsJohannes Ranke1-2/+2
Further relax two tests to pass build on Travis
2018-09-21Relax another test precision for travisJohannes Ranke1-1/+1
2018-09-21Reduce testing time, relax precision for travisJohannes Ranke1-15/+15
2018-09-21Skip long running test on travis to avoid timeoutJohannes Ranke1-5/+12
2018-09-21Improve fitting the two-component error modelJohannes Ranke2-40/+60
with respect to accuracy and robustness.
2018-09-20Test for convergence of error model to correct valuesJohannes Ranke1-0/+50
2018-09-20Fix testsJohannes Ranke2-2/+7
2018-09-20Skip irls test with two-component modelJohannes Ranke1-0/+1
as I just changed the algorithm
2018-09-18Use single core on travisJohannes Ranke1-1/+2
2018-09-18Support maximum twa for HS, test, docsJohannes Ranke1-18/+24
2018-09-15Restore test as we removed cor test, bump versionJohannes Ranke1-3/+3
2018-09-14Skip more tests on CRANv0.9.47.5Johannes Ranke5-46/+40
Static documentation rebuilt by pkgdown
2018-09-14Skip some more tests on CRANJohannes Ranke3-7/+6
Static documentation rebuilt by pkgdown
2018-09-14Remove two vignettes from the package but not from docsJohannes Ranke1-4/+4
- Rebuild static documentation - Adapt test to new approach to two component error model where the model is inadequate
2018-09-13Enable new error model in gmkinJohannes Ranke1-0/+4
Also improve the irls fitting of the error model and add a test for FOCUS_2006_C where the second component of the error model is zero
2018-02-06Skip some tests on CRAN, more quiet examplesv0.9.47.1Johannes Ranke3-40/+37
2018-01-30Reweighting with two-component error modelJohannes Ranke1-0/+46
Static documentation except articles rebuilt by pkgdown
2017-07-22Remove this test as results are platform dependentJohannes Ranke1-44/+0
On my box and winbuilder it was OK, but the CRAN checks show some different behaviour, so the results seem to be too ill-defined for testing for the type of warning. I had skipped this test earlier as it failed on r-forge.
2017-07-21Fix test_twa (correct "cores" argument")Johannes Ranke1-1/+1
2017-07-21Only use 1 core to pass test on windowsJohannes Ranke1-3/+3
2017-07-21Limit to one core for checks on windowsJohannes Ranke1-1/+1
2017-07-21Fix tests, try FOMC ill-defined againJohannes Ranke3-32/+34
2016-12-21TWA calculation for SFO, FOMC and DFOP (parent)Johannes Ranke1-0/+45
2016-06-28Skip testing for warning for ill-defined FOMC fitsJohannes Ranke1-0/+2
This caused the test suite to fail on r-forge for v0.9.43, obviously no warning was produced in one of the two expectations
2016-04-21Avoid unnecessary stop in edge case, improve warningsJohannes Ranke1-0/+42
mkinfit: Do not error out in cases where the fit converges, but the Jacobian for the untransformed model cost can not be estimated. Give a warning instead and return NA for the t-test results. summary.mkinfit: Give a warning message when the covariance matrix can not be obtained. A test has been added to containing such an edge case to check that the warnings are correctly issued and the fit does not terminate.
2015-06-21Do the t-test for untransformed parametersv0.9-36Johannes Ranke1-0/+28
2015-05-19Include output in README, use mkinsub() in example codeJohannes Ranke1-8/+9
2015-05-15Make mkinmod quiet in testsJohannes Ranke5-8/+8
2015-05-15Add synthetic data generated for UBA expertise, testsJohannes Ranke2-1/+68
The new tests are based on the results documented in the expertise I wrote for the UBA (currently unpublished).
2015-05-15Relax test tolerance for SFORB, FOCUS B and deSolveJohannes Ranke1-2/+2
While this was OK on my local windows machine, the test failed on winbuilder
2015-05-15Make FOMC test for FOCUS A robust towards non-convergenceJohannes Ranke1-5/+4
2015-05-15Also migrate test for mkinpredictJohannes Ranke1-28/+12
2015-05-15Skip testing FOMC with FOCUS A WindowsJohannes Ranke1-9/+15
Testing the package on winbuilder shows that the Port algorithm does not converge on Windows with this combination, which is not a problem as the FOMC model is not appropriate for the dataset anyways.
2015-05-15Finished migration from RUnit to testthatJohannes Ranke6-315/+363
2015-05-14Start of migration from RUnit to testthatJohannes Ranke4-0/+423

Contact - Imprint