mclogit - 0.9.7

2023-01-08: - Bugfix in 'predict.mmblogit' that caused an error if 'conditional=FALSE' was set.

2023-01-06: - More compact output of mblogit models random effects and diagonal covariance matrix.

2023-01-05: - Added support for alternative optimizers to be used in the inner iterations of MQL/PQL

- R
Published by melff over 3 years ago

mclogit -

2022-10-23: - Refactored MQL/PQL algorithm: Eliminated redundant code and adapted it to both 'nlm' and 'nlminb'

2022-10-16: - Fixed bug in MQL/PQL-objective function that led to false non-convergence and bias in variance parameter estimates

2022-10-12: - Support for starting values in random effects models - Support for restriction on random effects variances in multinomial baseline logit models

2022-10-09: - Improve handling of boundary values and singular information matrices

2022-10-07: - Remove spurious messages about missing starting values

2022-05-21: - Add checks of 'control=' argument of 'mclogit()' and 'mblogit()'.

- R
Published by melff over 3 years ago

mclogit -

Fixed bug in blockMatrix() and made it check for argument validity.

- R
Published by melff about 4 years ago

mclogit -

Hotfix of prediction method

- R
Published by melff about 4 years ago

mclogit -

2022-04-10: - Fix handling of singular initial covariance matrices in PQLMQL_innerFit - Issue a warning if models with random effects are compared using anova - Fix predict methods for mmclogit models - Handle DOIs in documentation as required by new guidelines

2022-01-16: - Fix prediction with complicated terms in the model - Add some more demos

2021-08-13: - predict.mmclogit: create W-Matrix only when really needed

2021-07-13: - Include variance parameters in the computation of degrees of freedom

2021-06-03: - Be less zealous about group-level covariates constant in some choice sets.

2021-05-30: - Added support for vertical-bar syntax for responses of conditional logit models.

- R
Published by melff about 4 years ago

mclogit -

• Added support for non-nested random effects.

- R
Published by melff about 5 years ago

mclogit -

• Fixed serious bug in the handling of multilevel random effects models.
• Detect some misspecified models with too many groups.

- R
Published by melff about 5 years ago

mclogit -

• Apply patch suggested by Ilya Yalchyk to improve handling of formula argument of mclogit() and mblogit().

- R
Published by melff over 5 years ago

mclogit -

This version is now published on CRAN

- R
Published by melff over 5 years ago

mclogit -

• summary() now shows the number of random effects groups.
• Support is added for the presentation of random effects variances and group numbers with mtable()

- R
Published by melff over 5 years ago

mclogit -

• Fixed prediction method also for mmclogit objects

- R
Published by melff over 5 years ago

mclogit -

• Refactored computations
• Fixed predictions from random-effects models where group indices are not a sequence of integers starting at 1.

- R
Published by melff over 5 years ago

mclogit -

This release fixes a typo in the mblogit() function.

- R
Published by melff over 5 years ago

mclogit - 0.8.6.4

Published now on CRAN.

- R
Published by melff over 5 years ago

mclogit -

This release fixes yet another bug in bMatTrns().

- R
Published by melff almost 6 years ago

mclogit -

Improved flexibility of prediction methods: "new" data frames do not need to contain exactly the same random effects identifiers as those used for model fitting, they may also contain a subset (as is useful for predmarg() from the "mpred" package.

- R
Published by melff almost 6 years ago

mclogit -

• This is a bugfix release that addresses re-opened issue #6

- R
Published by melff almost 6 years ago

mclogit -

This release improves the prediction methods from baseline-category and conditional logit models with random effects: If these models are fitted using the "PQL" technique, it is possible to obtain predictions conditional on the estimated random effects.

- R
Published by melff almost 6 years ago

mclogit -

• Fixed another silly bug in update.mclogit()

- R
Published by melff almost 6 years ago

mclogit -

- R
Published by melff almost 6 years ago

mclogit -

- R
Published by melff almost 6 years ago

mclogit -

Bugfix: Make 'update' work with missing dispersion= argument.

- R
Published by melff almost 6 years ago

mclogit -

Corrects remaining cross-ref issues in the documentation. This is the version currently on CRAN.

- R
Published by melff about 6 years ago

mclogit -

• Added and documented simulate() methods for "mblogit" and "mclogit" models.

- R
Published by melff about 6 years ago

mclogit -

• Improved documentation of estimation of baseline-logit models with overdispersion

- R
Published by melff about 6 years ago

mclogit -

• A "REML" variant is added to the PQL and MQL estimators for random-effects logit models.
• There is now a simulate() method for objects created by mclogit() or mblogit() (that do not involve random effects)

- R
Published by melff about 6 years ago

mclogit -

This is a bugfix release, addressing issue #7

- R
Published by melff about 6 years ago

mclogit - 0.8.2

• Added support for estimating (over-)dispersion parameters, based on techniques described in Afroz et al (2019)
• getSummary() methods are now explicitly declared in the NAMESPACE file. Package examples now also run on R 4.0.

- R
Published by melff about 6 years ago

mclogit - 0.8

• Support for the MQL technique of approxiate inference for random effects multinomial logit models is added.
• The algorithm for the computation of random effects multinomial logit model estimates is more stable, because estimates from the model variant without random effects are not used as starting values. (It appears paradox, but starting values from the no-random effects model variant created numerical instabilities in some instances.)
• The documentation now contains reference to the relevant literature, notably Agresti (2002), Breslow & Clayton (1993), and McFadden (1973).

- R
Published by melff about 6 years ago

mclogit - 0.7.2

• Fixes embarrassing bug that prevented estimation of models with more than two levels of random effects.
• Handles numerical difficulties more gracefully by giving users options to deal with them.

- R
Published by melff about 6 years ago

mclogit -

• The algorithm warns about and handles singular matrices in initial values of the inner iteration loop.
• The convergence criterium is based on the change in the linear part of the model instead of the deviance, like MASS::glmmPQL

- R
Published by melff about 6 years ago

mclogit - 0.7

• The algorithm has been completely refactored. It works much more along the lines of MASS::glmmPQL
• Internally a Cholesky-factor parametrisation is used to prevent negative variance estimates

- R
Published by melff about 6 years ago

mclogit - 0.6.3

• mblogit now handles responses with only two columns.
• mblogit now can deal with matrix responses that have rows that sum to zero.
• mclogit and mblogit now return a component named "df.residual" instead of "residual.df". (Fixes issue #4)

- R
Published by melff over 6 years ago

mclogit - 0.6.2

Merged PR from pmcharrison - typo fix fft -> fit

- R
Published by melff over 7 years ago

mclogit - 0.6.1

• Fixed matrix column selection in predict.mclogit if there is only one predictor (also PR from skyborla)

- R
Published by melff almost 8 years ago

mclogit -

• mclogit now complains about (i.e. throws an error exception) when the random effects structure cannot be estimated, e.g. because random effects are constant within choice sets and therefore drop out by the formation of conditional logits.

- R
Published by melff about 8 years ago

mclogit -

• Fixed some bugs in predict models for mclogit objects.
• Made sure that dummy coding is used for response factors even if they are ordinal

- R
Published by melff over 9 years ago

mclogit -

New features

• Support for random slopes in multinomial conditional logit models
• Support for random intercepts and random slopes in multinomial baseline logit models

- R
Published by melff over 9 years ago

mclogit -

• Fixed mclogit.fit() and predict.mclogit() to work better without covariates.
• Explicitely import package "methods"

- R
Published by melff over 9 years ago

mclogit -

• Import as from package "methods".
• Make sure nobs is defined in mclogit.fit.rePQL.
• Updated DESCRIPTION file: Maintainer email address changed and no "This package" at start of package descriptions.
• Fix display of number of observations
• Drop redundant coefficients

- R
Published by melff over 10 years ago

mclogit -

2015-08-01: - Added row and column names to estimator result of vcov() - Make sure that scripts run with "mclogit" loaded by devtools::load_all()

- R
Published by melff almost 11 years ago

mclogit -

2015-07-15: - mclogit, mclogit.fit: Added support for starting values.

2015-07-03: - predict.mblogit: 'contrasts.arg' not 'contast.arg' ... - predict-methods now should handle NAs in newdata arguments better.

2015-06-17: - Corrected handling of weights, and standard errors of prediction.

2015-06-15: - 'getSummary' methods now return "contrasts" and "xlevels" components. - Fixed prediction method for 'mclogit' results. - Added 'fitted' and 'predict' methods for 'mblogit' results.

2015-01-25: - Added support for multinomial baseline logit models in form of 'mblogit' as a frontend to 'mclogit.fit'

2015-01-23: - Added URLs to DESCRIPTION file

- R
Published by melff almost 11 years ago

mclogit - v0.3-2

- R
Published by melff over 11 years ago