Changes in flexmix version 2.3-8

  o The fit function in the M-step by default now is called with an
    argument containing the fitted component. This allows to use the
    parameter estimates from the previous step for initialization. Fit
    functions which do not require this now need a ... argument.
    Thanks to Hannah Frick and Achim Zeileis for requesting this
    feature.

  o Function initFlexmix() was added which is an alternative to
    stepFlexmix() where first several short runs of EM, SEM or CEM are
    performed followed by a long run of EM. 

Changes in flexmix version 2.3-7

  o A bug fixed in predict() and fitted() if a concomitant variable
    model is specified and aggregate = TRUE. Thanks to Julia Schiffner
    for pointing this out.

  o A bug fixed in FLXMRmgcv() if observations were removed in the
    M-step because their a-posterior probabilities were smaller than
    eps. Thanks to Ghislain Geniaux for pointing the problem out.

Changes in flexmix version 2.3-6

  o Vignettes moved from /inst/doc to /vignettes.

  o stepFlexmix() can now be called with a concomitant variable model
    FLXPmultinom() for k starting with 1 without getting an error. The
    concomitant variable model is internally replaced by
    FLXPconstant().

  o The boot() method for "flexmix" objects is extended to mixture
    models with concomitant variables and mixtures of linear mixed
    models.
	
  o A bug fixed in the summary() method for "flexmix" objects. The
    column post > 0 did not give the correct results if weights were
    used for fitting the mixture.

  o A bug fixed in the unique() method for "stepFlexmix" objects. This
    only occurred if components were dropped as well as if the EM
    algorithm did not converge for all repetitions. Thanks to
    Sebastian Meyer for pointing this out.

Changes in flexmix version 2.3-5

  o A bug fixed in posterior(). Fixed priors were always used, also if
    a concomitant variable model was present.

  o A method added for prior() such that if newdata is supplied and
    the object is of class "flexmix" the prior class probabilities for
    each observation are returned.
    
Changes in flexmix version 2.3-4

  o A generic method for nobs() is introduced in stats4 for R
    2.13.0. flexmix now does not define this generic function and
    logLik, AIC and BIC methods were modified to better exploit
    already available methods. Thanks to Prof. Brian D. Ripley for
    suggesting the modification.

Changes in flexmix version 2.3-3

  o A bug for boot() fixed for "flexmix" objects with an unbalanced 
    grouping variable. Thanks to Laszlo Sandor for pointing this out.

Changes in flexmix version 2.3-2

  o A bug for rflexmix() fixed for "flexmix" objects with a
    concomitant variable model. Thanks to Greg Petroski for pointing
    this out.

Changes in flexmix version 2.3-1

  o Functionality for bootstrapping finite mixture models added.

Changes in flexmix version 2.2-11

  o More generics and methods exported to use the refit method when
    extending flexmix in other packages.
  
Changes in flexmix version 2.2-10

  o For long formulas FLXMRglmfix() did not work properly due to the
    splitting of the formula into several parts by deparse(). This is
    fixed by pasting them together again. Thanks to Dustin Tingley for
    the bug report.

Changes in flexmix version 2.2-9
 
  o A new model driver FLXMRmgcv() is added which allows to fit
    regularized linear models in the components.

  o More generics and methods exported to allow extending flexmix in
    other packages.

Changes in flexmix version 2.2-8
 
  o The a-posteriori probabilities are now also determined as changed
    for FLXfit() for version 2.2-6 for refit().

  o Bug fixed for FLXfillconcomitant and refit when weights and
    grouping are present. A check was added that weights are identical
    within groups.

  o Function group() is now exported.

Changes in flexmix version 2.2-7
 
  o Bug in the FLXgetModelmatrix() method for the "FLXMRlmm" class
    fixed when determining identical random effects covariates for the
    grouping.

  o A new model driver for finite mixtures of linear mixed effects
    models with left-censored observations is added.

Changes in flexmix version 2.2-6
 
  o Determination of the a-posteriori probabilities made numerically
    more stable for small likelihoods. Thanks to Nicolas Picard for
    the code patch.

  o summary() for FLXMRmstep objects now returns similar output for
    which = "concomitant" as for flexmix version 2.0-1.

  o New demo driver FLXMCnorm1() for univariate Gaussian clustering.

  o Non-postive values for the maximum number of iterations for the
    "FLXcontrol" object are not valid. A validity check for this is
    now included.

Changes in flexmix version 2.2-5

  o Model class FLXMRfix introduced which is a subclass of FLXMR and a
    superclass for FLXMRglmfix which also extends FLXMRglm.

  o Model driver FLXMCfactanal added which allows to fit finite
    mixtures of Gaussian distributions where the variance-covariance
    matrix is estimated using factor analyzers.

  o Comparison of formulas now done using identical().

Changes in flexmix version 2.2-4

  o Model drivers FLXMRlmer() and FLXMRlmm() added for fitting finite
    mixtures of linear mixed effects models.

  o EIC() added as additional information criterion for assessing model fit.
	
  o Bug fixed in plot method for flexmix objects introduced in version 2.2-3.

Changes in flexmix version 2.2-3
	
  o New model driver FLXMCmvcombi() which is a combination of 
    Gaussian and binary. 

  o parameters() now also has a which argument in order to allow to
    access the parameters of the concomitant variable model.

  o Bug fixed in refit().

  o nobs() now returns the number of rows in the data.frame and not the
    number of individuals (similar as for example by lme).

Changes in flexmix version 2.2-0

  o vignette describing Version 2 added

  o isTRUE(all.equal()) replaced with identical().

  o Bug fixed for prior in flexmix().

  o New function relabel() to sort components (generic is in modeltools).

  o New example data generator ExLinear().

  o Fixed a bug in handling groups (gave an error for empty design matrices).

  o Added new model FLXMRrobglm() for robust estimation of GLMs.


Changes in flexmix version 2.1-0

  o Renamed cluster() to clusters() to avoid conflict with cluster() 
    from package survival

  o Bug fixed in internal functions using S4 generics and methods.


Changes in flexmix version 2.0-2

  o refit() now has a method argument. For method "optim" the
    variance-covariance matrix is determined using optim() to maximize the
    likelihood initialized in the solution found by the EM algorithm. 
    Method "mstep" refits the component specific and concomitant models 
    treating the posterior probabilities as given, i.e. performs an M-step 
    of the EM algorithm.


Changes in flexmix version 2.0-1

  o Lapply() added which allows to apply a function to each component
    of a finite mixture

  o KLdiv() modified to allow for determination with a discrete and a
    continuous version of the KL divergence


Changes in flexmix version 2.0-0

  o Model driver for zero-inflated component specific models.

  o Latent class analysis for binary multivariate data is now
    possible to estimate for truncated data where the number of 
    observations with pattern only zeros is missing.

  o new argument newdata for cluster()

  o new unique() method for "stepFlexmix" objects


Changes in flexmix version 1.9-0

  o New class definitions for component specific models and
    concomitant variable models.

  o fitted() and predict() now have an aggregate argument in order to
    be able to determine the aggregated values over all components.

  o The package has now a vignette presenting several applications of
    finite mixtures of regression models with varying and fixed effects 
    on artificial and real data which can be a accessed using the
    command vignette("regression-examples").

  o The vignette "flexmix-intro" was adapted to reflect the changes
    made in the package.

  o stepFlexmix() now returns an object of class stepFlexmix which has
    a print and plot method. In addition getModel() can be used to
    select an appropriate model.

  o flexmix() now has a weights argument for multiple identical
    observations.

  o New model drivers for latent class analysis with Bernoulli and
    Poisson distributed multivariate observations.

  o Variants of the EM algorithm have been modified to correspond to
    CEM and SEM. These names can now also be used for specifying the
    classify slot of the FLXcontrol object.


Changes in flexmix version 1.8-1

  o The package can now fit concomitant variable models.

  o New M-step driver for regression models with varying and fixed
    effects.
    
  o ICL information criterion


Changes in flexmix version 1.1-2

  o Fixed a bug that made the log-likelihood infinity for observations
    where all posteriors are numerically zero

  o Fixed a bug for formulae with dots.

  o posterior() now has a newdata argument.

  o New demo driver for model-based clustering of binary data.
 
  o Adapted to changes in summary.glm() of R version 2.3.0.


Changes in flexmix version 1.1-1

  o The 'cluster' argument of flexmix() may now also be a matrix of 
    posterior probabilities.

  o Fixed a bug to make size table work in case of empty clusters.

  o Fixed a bug in likelihood computation for grouped observations.

  o The artificial NPreg data now also have a binomial response, 
    added example to help("flexmix").


Changes in flexmix version 1.1-0

  o The FLXglm driver now has an offset argument.

  o New data set seizure as example for a Poisson GLM
    with an offset.

  o fitted() can be used to extract fitted values from flexmix
    and FLXrefit objects.

  o New accessor methods cluster() and posterior().

  o The package now uses lazy loading and has a namespace.


Changes in flexmix version 1.0-0

  o The package has now an introductionary vignette which can be
    accessed using the command vignette("flexmix-intro"). The vignette
    has been published in the Journal of Statistical Software, Volume
    11, Issue 8 (www.jstatsoft.org), and the paper should be used as
    citation for flexmix, run citation("flexmix") in R 2.0.0 or newer
    for details.

  o Several typos in help pages have been fixed.
  
	
Changes in flexmix version 0.9-1

  o Adjust for R 2.0.0.

  o Fixed a bug in the summary and plot methods of flexmix objects in
    case of empty clusters.

  o stepFlexmix takes two new arguments: `compare' allows fo find
    minimum AIC/BIC solutions in addition to maximum likelihood,
    `verbose' gives some information about progress.

  o Use a default of verbose=0 in FLXcontrol (better in combination 
    with stepFlexmix). 

	
Changes in flexmix version 0.9-0

  o new summary() and plot() methods for flexmix objects

  o FLXglm objects can now be automatically refit()ted to get table of
    significance tests for coefficients

  o new function stepFlexmix() for more automated model search

  o the artificial example data now have functions to
    create them and a pre-stored data sets, new function plotEll() to
    plot 2d Gaussians with confidence ellipses.

  o new function KLdiv() to compute Kullback-Leibler divergence of
    components

  o the calculation of the degrees of freedom for FLXmclust was wrong


Changes in flexmix version 0.7-1

  o fixed some codoc problems (missing aliases)


First version released on CRAN: 0.7-0