Changes in version 1.2.6 (2025-06-02) o added data_SRHS_long.rda to inst/extdata/ Changes in version 1.2.5 (2024-08-30) o added Authors@R o fixed example 3 in vignette_mvord2.Rmd o in marginal_predict(), predict() and joint_probabilities() newdata can now contain no response variable; o adjusted docu for joint_probabilities() Changes in version 1.2.4 (2024-07-07) o change handling of data argument to work also with tibbles Changes in version 1.2.3 o fixed bug in transf_thresholds_fixall Changes in version 1.2.2 (2024-02-27) o changed maintainer Changes in version 1.2.1 (2024-01-13) o fixed bugs from the change NCOL(NULL) returns 0 rather than 1 Changes in version 1.2.0 o added essay_data to mvord datasets o added a combis argument in mvord.control which allows the specification of the combinations to be used, if user does not want to use all. o for logit link replaced mnormt::sadmvt with mvtnorm::pmvt. Function sadmvt cannot not handle dimension larger than d=20 o fixed bug for standard error when threshold.coef are e.g., rep(1:3, 3). Changes in version 1.1.1 (2021-03-17) o fixed bug for value0 with cor_ar1 Changes in version 1.1.0 (2020-11-17) o speed up in PL_fun - minimal changes in the 10+ decimal places for some specific lead to minimal deviations in the optimization results of the general purpose optimizers. In some cases more evaluation steps are needed. o implementation of PL_fast o fixed bug in vector constraints for coef Changes in version 1.0.2 o fixed bug in prediction with categorial variables with newdata Changes in version 1.0.1 (2020-05-28) o changes in checks due to new treatment when combining ordered factors in R 4.0.1. Changes in version 1.0.0 (2020-04-17) o publication in Journal of Statistical Software Changes in version 0.3.6 (2019-09-22) o fixed bug in thresholds checks for two binary responses Changes in version 0.3.5 (2019-03-06) o added suppressWarnings(RNGversion("3.5.0")) in tests Changes in version 0.3.4 (2019-02-20) o two additional arguments have been added to the error structures: value = numeric(0) (for initial values) and fixed = FALSE (for allowing the user to fix the parameters of the error structure to the values specified in argument value) o efficiency improvements by using rep.int, paste0, seq_len, seq_along where possible Changes in version 0.3.3 (2018-10-31) o fixed bug standard errors with threshold.constraints Changes in version 0.3.2 (2018-10-03) o function polycor() for polychoric correlations is now available o added packages minqa, BB, ucminf and dfoptim as dependencies Changes in version 0.3.1 (2018-06-12) o out-of-sample predictions are now available o additional examples are now available in an additional vignette Changes in version 0.3.0 (2018-05-03) o new model design with multiple measurement objects MMO and MMO2 (former mvord2()) o removed function mvord2() o removed argument scale (covariates are scaled internally for optimizer) o fixed bug in standard errors with binary outcomes o removed argument response.names (if specific ordering is desired this can be performed by an ordered factor for the multiple measurement index) o additional input types are now applicable for ordinal response variables o fixed bug trace o control = mvord.control() o additional checks o new function name marginal_predict() (instead of marginal.predict()) o new function name joint_probabilities() (instead of get.prob()); type "class" instead of "class.max" o new function name error_structure() (instead of get_error_struct()) Changes in version 0.2.1 (2017-11-29) o adapted predict(), get.prob(), marginal.predict() o added contrasts as argument to mvord() and mvord2() o fixed bug in VGAM design of coef.constraints o changed internal design matrices Changes in version 0.2.0 (2017-11-10) o implemented nobs.mvord() o implemented vcov.mvord() o implemented terms.mvord() o implemented model.matrix.mvord() o implemented fitted.mvord() o implemented logLik.mvord() o AIC() and BIC() are now available o additional argument scale o new (additional) design for coef.constraints in analogy to constraints in VGAM o category-specific regression parameters o additional argument offset o new class "mvlink" o implemented coef.constraints() o logPL(), claic() and clbic() are not exported anymore. use logLik(), AIC() and BIC() instead. o renamed cor_general, cor_ar1, cor_equi, cov_general o implemented names_constraints() Changes in version 0.1.0 (2017-10-17) o changed function name from multord() to mvord() o changed function name from multord2() to mvord2() o implemented predict function() o implemented predict.marginal function() o implemented get.prob function() o changed link probit to mvprobit() o changed link logit to mvlogit(df = 8L); default value of degrees of freedom is 8 o implemented a more flexible multivariate logistic distribution with logistic marginals and t copula (df are settable) o additional argument PL.lag