47 lines (33 loc) · 1.53 KB

varian 0.3.0

Major Changes

parallel_stan() has been removed as the rstan sampling function can run in parallel now. To use multiple cores now, follow the rstan approach of: rstan_options(auto_write = TRUE) options(mc.cores = 4) if you wanted 4 cores, for example,
varian() now only requires a single seed to be set, as this is now controlled by rstan rather than the removed parallel_stan() function.

New Features

varian() can now include quadratic effects of latent means and intraindividual variabilities using the new arguments, UQ = TRUE and IIVQ = TRUE.
summary.vm() method now added for a convenient summary.
shinystan package added as a suggested package. This implements interactive and high quality model diagnostics. This will likely replace the vm_diagnostics() function in the near future.

varian 0.2.0

Major Changes

vm_predict() renamed to varian() reflecting a unification of separate functions into a more general purpose, variability analysis function.

New Features

varian() now allows different model designs including "V" to estimate intra-individual variability alone (without using it as a predictor) and "V -> M -> Y" to estimate a simple mediation model.
Many back end changes including more pre-modeling data checks and better estimates for start values.

varian 0.1.0

Functions

vm_predict() calculates the intraindividual variability and uses this to predict an outcome