Changes in version 0.3.5 (2025-10-27)                  

    o   Replaced qr() by lm.wfit() to fix memory issue

    o   Removed acceleration scheme from fitting algorithm

    o   Updated maintainer email address, removed CPP11 restriction,
	and arxiv urls to doi (#24 @zeileis)

                 Changes in version 0.3.4 (2022-08-10)                  

    o   Added vcov.APEs() generic to extract the covariance matrix
	after getAPEs().

    o   Improved the finite sample performance of bias corrections for
	the average partial effects in case of perfectly classified
	observations.

    o   Bias corrections for the average partial effects, i.e.
	getAPEs() after biasCorr(), do not require an offset algorithm
	anymore.

    o   The default option 'n.pop' in getAPEs() has been changed. Now
	the estimated covariance consists of the delta method part
	only, i.e. correction factor = 0.

    o   Improved the numerical stability of the bias corrections.

    o   biasCorr() now also supports one-way fixed effects models.

    o   Added bias corrections for 'cloglog' and 'cauchit'.

    o   feglm() and feglm.nb() do not return a matrix of scores
	anymore. Instead they, optionally, return the centered
	regressor matrix. The corresponding option in feglmControl() is
	'keep.mx'. Default is TRUE.

    o   Improved the numerical stability of the step-halving in
	feglm().

    o   Changed the projection in the MAP algorithm.

    o   The default option 'center.tol' in feglmControl() has been
	lowered to better handle fitting problems that are not
	well-behaved.

    o   Added optional 'weights' argument to feglm() and feglm.nb().

    o   Updated documentation.

                 Changes in version 0.3.3 (2020-10-30)                  

    o   Stopping condition of feglm.nb() has been adjusted to better
	match that of glm.nb().

    o   feglm.nb() now additionally returns 'iter.outer' and
	'conv.iter' based on iterations of the outer loop. Previously
	'iter' and 'conv' were overwritten.

    o   Step-halving in feglmFit() and feglmOffset() is now similar to
	glm.fit2().

    o   Fixed an error in the covariance (influence function) of
	getAPEs().

    o   Updated some references in the documentation and vignette.

    o   Fixed some typos in the documentation and vignette.

                 Changes in version 0.3.2 (2020-01-12)                  

    o   Added option 'panel.structure' to biasCorr() and getAPEs().
	This option allows to choose between the two-way bias
	correction suggested by Fernández-Val and Weidner (2016) and
	the bias corrections for network data suggested by Hinz,
	Stammann, and Wanner (2020). Currently both corrections are
	restricted to probit and logit models.

    o   Added option 'sampling.fe' to getAPEs() to impose simplifying
	assumptions when estimating the covariance matrix.

    o   feglm() now permits to expand functions with poly() and bs()
	(#9 @tcovert).

    o   feglm() now uses an acceleration scheme suggested by Correia,
	Guimaraes, and Zylkin (2019) that uses smarter starting values
	for centerVariables().

    o   Added an example of the three-way bias correction suggested by
	Hinz, Stammann, and Wanner (2020) to the vignette.

    o   The control parameter 'trace' now also returns the current
	parameter values as well as the residual deviance.

    o   Fixed an error in getAPEs() related to the estimation of the
	covariance.

    o   Fixed a bug in the internal function that is used to estimate
	spectral densities.

                 Changes in version 0.3.1 (2019-05-24)                  

    o   All routines now use setDT() instead of as.data.table() to
	avoid unnecessary copies (suggested in #6 @zauster).

    o   feglm.nb() now returns 'iter' and 'conv' based on iterations of
	the outer loop.

    o   Fixed a bug in feglm() that prevented to use I() for the
	dependent variable.

    o   Fixed an error in getAPEs() related to the covariance.

    o   The last line of print.summary.feglm() now ends with a line
	break (#6 @zauster).

    o   The internal function feglmFit() now correctly sets 'conv' if
	the algorithm does not converge (#5 @zauster).

    o   Fixed some typos in the vignette.

                  Changes in version 0.3 (2019-05-14)                   

    o   feglm() now allows to estimate binomial model with fractional
	response.

    o   Added feglm.nb() for negative binomial models.

    o   Added post-estimation routine biasCorr() for analytical
	bias-corrections (currently restricted to logit and probit
	models with two-way error component).

    o   Added post-estimation routine getAPEs() to estimate average
	partial effects and the corresponding standard errors
	(currently restricted to logit and probit models with two-way
	error component).

    o   getFEs() now returns a list of named vectors. Each vector
	refers to one fixed effects category (suggested in #4
	@zauster).

    o   Changed stopping condition to the one used by glm().

    o   Changed least squares fit to QR (similar to lsfit() used by
	glm()).

    o   Source code and vignettes revised.

                  Changes in version 0.2 (2018-07-31)                   

    o   Initial release on CRAN.