mkin/DESCRIPTION, branch v0.9.49.5 Fitting kinetic models to chemical degradation data (also on github) https://cgit.jrwb.de/mkin/atom?h=v0.9.49.5 2019-07-04T06:10:33Z Address failures of CRAN checks, improve NEWS 2019-07-04T06:10:33Z Johannes Ranke jranke@uni-bremen.de 2019-07-04T06:10:33Z urn:sha1:7e8d788d298b8e1492fd8f62d88456e99e0f5992 Static documentation rebuilt by pkgdown Update DESCRIPTION, improve NEWS 2019-07-03T14:45:09Z Johannes Ranke jranke@uni-bremen.de 2019-07-03T14:45:09Z urn:sha1:798195d367111e592e5247309b90fac29ed7ddfd Static documentation rebuilt by pkgdown Additional algorithm "d_c", more tests, docs 2019-06-04T13:09:28Z Johannes Ranke jranke@uni-bremen.de 2019-06-04T13:09:28Z urn:sha1:95178837d3f91e84837628446b5fd468179af2b9 The new algorithm tries direct optimization of the likelihood, as well as a three step procedure. In this way, we consistently get the model with the highest likelihood for SFO, DFOP and HS for all 12 new test datasets. Fix a bug that could lead to wrong chi2 error levels 2019-05-23T17:19:27Z Johannes Ranke jranke@uni-bremen.de 2019-05-23T17:19:27Z urn:sha1:db472848915a8065347b8f81477a86928b89fbcf After the OLS step, use OLS parameter estimates 2019-05-07T06:12:27Z Johannes Ranke jranke@uni-bremen.de 2019-05-07T06:12:27Z urn:sha1:8a3475c59f3d91ce5ce7d980d6de09360617e7fe - Fix the respective error in the code - Static documentation rebuilt by pkgdown Adapt tests, vignettes and examples 2019-04-10T08:17:35Z Johannes Ranke jranke@uni-bremen.de 2019-04-10T08:17:35Z urn:sha1:194659fcaccdd1ee37851725b8c72e99daa3a8cf - Write the NEWS - Static documentation rebuilt by pkgdown - Adapt mkinerrmin - Fix (hopefully all) remaining problems in mkinfit Direct error model fitting works 2019-04-04T15:21:13Z Johannes Ranke jranke@uni-bremen.de 2019-04-04T13:42:23Z urn:sha1:7a1d3d031aa23fce723ac4f4c8e4bb5d64959447 - No IRLS required - Removed optimization algorithms other than Port - Removed the dependency on FME - Fitting the error model 'obs' is much faster for the FOCUS_2006_D dataset and the FOMC_SFO model (1 second versus 3.4 seconds) - Vignettes build slower. Compiled models needs 3 minutes instead of 1.5 - For other vignettes, the trend is less clear. Some fits are faster, even for error_model = "const". FOCUS_Z is faster (34.9 s versus 44.1 s) - Standard errors and confidence intervals are slightly smaller - Removed code for plotting during the fit, as I hardly ever used it - Merged the two cost functions (using transformed and untransformed parameters) into one log-likelihood function Set URL to documentation page 2019-03-19T15:01:00Z Johannes Ranke jranke@uni-bremen.de 2019-03-19T15:01:00Z urn:sha1:7e643cf3585be1e4fb758c6bf40e807616767f5a Review tests and examples based on winbuilder results 2019-03-04T09:12:43Z Johannes Ranke jranke@uni-bremen.de 2019-03-04T09:12:43Z urn:sha1:d0a358a8f4fd03c95536409523e492eb15726f44 Static documentation rebuilt by pkgdown Fix CAKE export test to actually work 2019-02-27T07:06:59Z Johannes Ranke jranke@uni-bremen.de 2019-02-27T07:06:59Z urn:sha1:c446b59e675aeff08ff7205b05f06cd81bf6dc68 Also test the model specification via the link argument