                        Changes in version 0.7-4                        

    o   Fixed tests failing on macOS.

                 Changes in version 0.7-3 (2026-10-16)                  

    o   C-Routines are now symbols prefixed with C_ (e.g.,
	C_wght_LL_grad_alternative).

    o   .onUnload now also unloads the DLL.

    o   Changed call to structure() from the deprecated special name
	.Dim to dim.

                 Changes in version 0.7-2 (2025-05-30)                  

    o   Implemented CRAN-Notes.

                 Changes in version 0.7-1 (2021-05-18)                  

    o   summary.DirichletRegModel now returns a
	summary_DirichletRegModel-class object without printing it (see
	<https://github.com/maiermarco/DirichletReg/issues/3>).

    o   Moved rmarkdown to Suggests to comply with new version of knitr
	(see <https://github.com/yihui/knitr/issues/1864>).

                 Changes in version 0.7-0 (2020-05-29)                  

    o   Fixed: Cutoff in formula-deparsing removed.

    o   Removed rgl dependency (package will be loaded if necessary).

    o   Moved to GitHub.

                 Changes in version 0.6-3 (2015-11-12)                  

    o   Instead of producing an error, density functions now return NaN
	with a warning if any element in alpha is <= 0.

                 Changes in version 0.6-2 (2015-02-09)                  

    o   Fixed an error in the ternary plot (bottom axis tick labels
	were printed in reversed order; pointed out by Emilio A. Laca).

    o   Imports functions from default packages, as required by the new
	CRAN-check.

    o   Changed the Description field to eliminate the note “Malformed
	Description field: should contain one or more complete
	sentences.” when checking the package.

                 Changes in version 0.6-1 (2014-12-10)                  

    o   Re-derived gradients/Hessians, optimized C-code, and tweaked
	computation of starting values which leads to considerably
	better performance.

    o   Fixed some bugs in drop1.

    o   Expanded testthat checks.

                        Changes in version 0.6-0                        

    o   All likelihood and gradient functions are now written in C (all
	called via .Call() now instead of .C()) which leads to a
	considerable gain in speed.

    o   A drop1 method for Dirichlet regression models was added.  As
	it is still experimental and will probably change, use it with
	care.  Options such as scope will be added in one of the next
	releases.

    o   Fixed a bug in the print method for confidence intervals and
	one in confint.DirichletRegModel().

    o   Expanded the testthat check- and test-suite.

    o   Known issue: If you have collinear (aliased) terms, the
	estimation will fail.  This will be handled automatically in
	subsequent releases, but for now, please remove the respective
	terms.  If you fit a model and it says something like: Error in
	prepareFixed(start = start, activePar = NULL, fixed = fixed):
	At least one parameter must not be fixed using argument 'fixed'
	you most likely have collinear terms or “empty” combinations of
	interaction terms.

                 Changes in version 0.5-2 (2014-09-18)                  

    o   Fixed checking functions in tests/testthat.

                 Changes in version 0.5-1 (2014-09-18)                  

    o   Fixed a bug when using the subset argument.

    o   Added tolerance for normalization check to DR_data.

    o   moved NEWS to the new fancy NEWS.Rd file/format.

    o   Added the possibility to do quick analyses and transforming
	data “on the fly”, like DirichReg(DR_data(A) ~ 1).  However
	this is only intended for quick checking purposes and may be
	removed at any time.

                        Changes in version 0.5-0                        

    o   Transformation in DR_data is now not only TRUE/FALSE, but, by
	default, a small numeric value to avoid troubles with floating
	point numbers close to 0 or 1.

    o   Time-critical routines were implemented in C (pure R versions
	are available, see ?ddirichlet).

    o   anova.DirichletRegModel now invisibly returns results as an
	object that is printed by a method.

    o   Optimized estimation routines.

    o   Fixed a bug in the predict method.

    o   Started development of a comprehensive test-suite using
	testthat.

    o   Published a working paper on the package:
	Maier, M. J. (2014). _DirichletReg: Dirichlet Regression for
	Compositional Data in R._ Research Report Series / Department
	of Statistics and Mathematics, 125. WU Vienna University of
	Economics and Business, Vienna.
	doi:10.57938/ad3142d3-2fcd-4c37-aec6-8e0bd7d077e1
	<https://doi.org/10.57938/ad3142d3-2fcd-4c37-aec6-8e0bd7d077e1>

    o   Added vignette with code to the working paper.

    o   Added citation info.

                        Changes in version 0.4-1                        

    o   The trafo Argument of DR_data has been changed, because it has
	lead to problems in practical applications when numbers very
	close to 0 or 1 were present.

    o   DR_data checks for negative values and generates an appropriate
	error message.

    o   DR_data has been made more robust in the presence of NAs.

                 Changes in version 0.4-0 (2012-09-06)                  

    o   Data structure generated by DR_data has changed – the new
	objects can now be integrated into data frames.

    o   Formula processing is now handled by the package Formula.

    o   New methods have been implemented, especially for the class
	DirichletRegModel.

    o   The documentation is now quite complete.

    o   Some speed improvements could be achieved.

    o   Lots of minor (invisible) changes.

                        Changes in version 0.002                        

    o   Added the analytical Gradient and Hessian for both
	parametrizations.

    o   Optimization: preliminary results by BFGS that become starting
	values for Newton-Raphson optimization computing the final
	results.

    o   Implemented some residuals

    o   Updated help entries
