aboutsummaryrefslogtreecommitdiff
path: root/man/mkinfit.Rd
AgeCommit message (Collapse)AuthorFilesLines
2020-05-29Warn if standardized residuals are unlikely normalJohannes Ranke1-63/+0
This revealed a bug in the data returned in mkinfit$data in the case of the d_3 algorithm, which also affected the residual plot - the data from the direct fitting was not returned even if this was the better method.
2020-05-15Use markdown links in mkinfit documentationJohannes Ranke1-24/+24
2020-05-12Another documentation updateJohannes Ranke1-11/+8
2020-05-12Improve documentation, now using a spell checkerJohannes Ranke1-1/+1
2020-05-12Documentation improvements, rebuild static docsJohannes Ranke1-56/+62
2020-05-09Avoid the call to merge for analytical solutionsJohannes Ranke1-6/+0
This increases performance up to a factor of five!
2020-05-08Slight performance gain by reducing default n.outtimesJohannes Ranke1-2/+2
2020-05-06Change implementation of analytical solutionsJohannes Ranke1-4/+4
Preparing for symbolic solutions for more than one compound
2020-03-30First nlme fits for models with a metaboliteJohannes Ranke1-9/+24
2019-10-31Add likelihood ratio test and other methods, fixesJohannes Ranke1-1/+3
The likelihood ratio test method is lrtest, in addition, methods for update and residuals were added.
2019-10-25Use roxygen for functions and methodsJohannes Ranke1-232/+206
2019-10-22Go back to d_3 for errmod = "obs"Johannes Ranke1-3/+2
in the hope that this makes plotting cross-platform also for this error model
2019-10-21Refactor mkinfit, infrastructure workJohannes Ranke1-10/+8
mkinfit objects now include an ll() function to calculate the log-likelihood. Part of the code was refactored, hopefully making it easier to read and maintain. IRLS is currently the default algorithm for the error model "obs", for no particular reason. This may be subject to change when I get around to investigate. Slow tests are now in a separate subdirectory and will probably only be run by my own Makefile target. Formatting of test logs is improved. Roundtripping error model parameters works with a precision of 10% when we use lots of replicates in the synthetic data (see slow tests). This is not new in this commit, but as I think it is reasonable this closes #7.
2019-07-09Adaptations for gmkinJohannes Ranke1-3/+8
Address winbuilder check problems, update check log, update of static docs
2019-07-05Static documentation rebuilt by pkgdownJohannes Ranke1-1/+2
2019-07-02TypoJohannes Ranke1-1/+1
2019-06-04Additional algorithm "d_c", more tests, docsJohannes Ranke1-0/+40
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.
2019-05-02Better initials for error model parametersJohannes Ranke1-0/+7
- Also make it possible to specify initial values for error model parameters. - Run tests - Rebuild docs
2019-04-10Adapt tests, vignettes and examplesJohannes Ranke1-126/+51
- Write the NEWS - Static documentation rebuilt by pkgdown - Adapt mkinerrmin - Fix (hopefully all) remaining problems in mkinfit
2019-04-04Direct error model fitting worksJohannes Ranke1-29/+17
- 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
2018-11-23Add logLik method to enable AIC() on mkinfit modelsJohannes Ranke1-1/+4
Further relax two tests to pass build on Travis
2018-09-13Enable new error model in gmkinJohannes Ranke1-2/+3
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
2018-09-06Small doc corrections and improvementsJohannes Ranke1-7/+8
2018-07-18Correct references to the Rocke and Lorenzato modelJohannes Ranke1-3/+6
Rename 'sigma_rl' to 'sigma_twocomp' as the Rocke and Lorenzato model assumes lognormal distribution for large y. Rebuild static documentation.
2018-01-30Reweighting with two-component error modelJohannes Ranke1-6/+22
Static documentation except articles rebuilt by pkgdown
2016-11-18Improve examples for showing with pkgdownJohannes Ranke1-13/+15
- Use quiet= TRUE in dontrun sections - Use mkinsub in model definitions - Avoid \code{\link{}} in titles
2016-11-18Static documentation rebuilt by pkgdown::build_site(run_dont_run = TRUE)Johannes Ranke1-2/+2
Using branch 'jranke' on jranke/pkgdown, fixing issues hadley/pgkdown#178 and hadley/pkgdown#213 Remove plot=TRUE from mkinfit calls also in dontrun sections of examples to avoid a flood png files documenting the progress of the fit.
2016-11-17Remove trailing whitespace, clean headersJohannes Ranke1-29/+29
Also ignore test.R in the top level directory, as it is not meant to be public
2016-06-27Improve listing of weights in output, update NEWSJohannes Ranke1-6/+5
2016-03-23Added 'from_max_mean' to 'mkinfit'.Johannes Ranke1-0/+8
- Now it is easier to do decline kinetics from the maximum occurrence - Static documentation rebuilt by staticdocs::build_site() - Fixed an URL in the README, thanks to the win-builder check with R-devel
2015-07-22Remove periods in help file titlesJohannes Ranke1-1/+1
2015-06-29Quiet fitting in examples and more links for static docs.Johannes Ranke1-0/+7
2015-06-23Small documenation changesJohannes Ranke1-1/+1
2015-06-20Low-level generation of compiled modelsJohannes Ranke1-1/+1
As it is unclear if and when ccSolve will be published on CRAN, the generation, compilation and use of the C version of the system of differential equations was developed for mkin, inspired and guided by the code from the ccSolve package. Many thanks again to Karline Soetaert for all of her work on this and other R packages. Now all model types, including the Hockey-Stick model for the parent compund and the IORE model for parent and/or metabolites can be compiled.
2015-06-05Make the examples that are run more quietJohannes Ranke1-2/+2
2015-05-19Include output in README, use mkinsub() in example codeJohannes Ranke1-14/+14
2015-05-18Make checking faster, testing is done by testsJohannes Ranke1-13/+6
2015-05-15Merge branch 'master' into compile_odesJohannes Ranke1-0/+1
2015-05-15Reduce time to run mkinfit exampleJohannes Ranke1-0/+2
2015-04-18Add a vignette showing the performance of compiled modelsJohannes Ranke1-6/+10
2015-04-16Update copyright header, improve documentation of solution_typeJohannes Ranke1-2/+3
2015-04-16Check compiled mkinmod versions in examples and testsJohannes Ranke1-3/+18
Tests all pass (Linux)
2015-04-16Correctly close braces of examples sectionJohannes Ranke1-0/+1
2015-04-14Compile differential equation models with ccSolve packageJohannes Ranke1-2/+3
If the ccSolve package is available, and time is not in the right hand side of the equations (i.e. if only SFO and SFORB models are used), the differential equation model is compiled from auto-generated C code. Currently, one test (FOCUS 2006 D SFO_SFO) fails
2014-11-12Merge branch 'master' of https://github.com/jranke/mkin into ioreJohannes Ranke1-1/+1
2014-11-12IORE working for parent and metabolites, introduce mkinsubJohannes Ranke1-0/+6
2014-11-12Fix the typo in the help page, tooJohannes Ranke1-1/+1
2014-10-15Switch to using the Port algorithm per defaultJohannes Ranke1-12/+12
2014-10-14Add a comment about transforming fractionsJohannes Ranke1-4/+10
2014-07-24Improve formulationJohannes Ranke1-1/+1

Contact - Imprint