NEWS.md

In remiod: Reference-Based Multiple Imputation for Ordinal/Binary Response

remiod 1.0.1

New features

• Ordinal Probit Model: new model option under algorithm='jags'. The option can be requested with argument models = 'opm' (or named vectors, for example, models = c(y1='opm',y2='opm')).
• algorithm='tang_seq': computation is speed up.
• get_subset: function is exported for extracting specified set of parameters from MCMC samples. The extracted samples have a class of 'mcmc', and can be directly used by functions/packages which take a 'mcmc' object.

Minor improvements and bug fixes

• Bug fix: when using a J2R imputation of continuous outcome through GLM with family = gaussian() using JAGS.

remiod documentation built on Nov. 19, 2022, 1:06 a.m.